This is a partial publication list. Please find the complete list in my cv.
Selected Publications
Automated Negotiation
Mohammad, Y. (2026). Automated Negotiation with No Information about Partner Utility Functions Using the Tentative Acceptance Unique Offers Protocol. Autonomous Agents and Multi-Agent Systems, 40(1), 21.
@article{mohammad2026tau,
keywords = {journal, important},
focus = {negotiation},
author = {Mohammad, Yasser},
title = {Automated Negotiation with No Information about Partner Utility Functions Using the Tentative Acceptance Unique Offers Protocol},
journal = {Autonomous Agents and Multi-Agent Systems},
volume = {40},
number = {1},
pages = {21},
year = {2026},
doi = {10.1007/s10458-026-09745-9}
}
With the widespread adoption of AI in industrial and business operations, finding methods to reach agreement between intelligent agents representing self-interested entities in a general-sum environment (cooperation within competition) is attracting more interest in the research community. Negotiation is a common process for reaching agreements between people and human institutions. Automated negotiation is thus being considered for cooperation within competition situations involving AIs. The most widely used protocols for automated negotiation are the Stacked Alternating Offers Protocol (SAOP) for multilateral negotiations and the Alternating Offers Protocol (AOP) for bilateral negotiations which directly model bargaining as in human marketplaces. Several strategies have been proposed for these protocols over the years. In this paper, we propose a modification of the AOP and SAOP protocols and a method for adapting negotiation strategies to the new protocol. We show empirically that the proposed approach leads to higher expected advantage for all agents, and achieves higher agreement rate, higher welfare, and fairer agreements faster. This is achieved at the expense of a small increase in information revelation.
Automated Negotiation
Mohammad, Y. (2023). Optimal Time-Based Strategy for Automated Negotiation. Applied Intelligence, 53(6), 6710–6735.
@article{mohammad2023optimaltime,
keywords = {journal, important},
focus = {negotiation},
author = {Mohammad, Yasser},
title = {Optimal Time-Based Strategy for Automated Negotiation},
journal = {Applied Intelligence},
volume = {53},
number = {6},
pages = {6710--6735},
year = {2023},
doi = {10.1007/s10489-022-03662-6}
}
Recent years are showing increased adoption of AI technology to automate business and production processes thanks to the recent successes of machine learning techniques. This leads to increased interest in automated negotiation as a method for achieving win-win agreements among self-interested agents. Research in automated negotiation can be traced back to the Nash bargaining game in the mid 20 th century. Nevertheless, finding an optimal negotiation strategy against an unknown opponent with an unknown utility function is still an open area of research. The most recent result in this area is the Greedy Concession Algorithm (GCA) which can be shown to be optimal under specific constraints on both the negotiation protocol (non-repeating offers), opponent (static acceptance-model) and search space (deterministic time-based strategies). In this paper, we extend this line of work by providing an algorithmically faster version of GCA called Quick GCA which reduces the time-complexity of the search process from O( 2 ) to O( ) where is the size of the outcome-space and is the number of negotiation rounds allowed. Moreover, we show that GCA/QGCA can be applied in a more general setting; Namely with repeating-offers protocols and to search the more general probabilistic time-based strategies. Finally, we heuristically extend QGCA to more general opponents with general time-dependent acceptance-model and negotiation settings (real-time limited negotiations) in three steps called , , and that iteratively and greedily modify the policy proposed by applied to an approximate static acceptance model . The paper evaluates the proposed approach empirically against state of the art negotiation strategies (winners of all relevant ANAC competition winners) and shows that it outperforms them in a wide variety of negotiation scenarios.
Automated Negotiation
Mohammad, Y., & Nakadai, S. (2022). Concurrent Negotiations with Global Utility Functions. Proceedings of the 21st International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2022), 1947–1949. https://www.ifaamas.org/Proceedings/aamas2022/pdfs/p1947.pdf
@inproceedings{mohammad2022concurrentglobal,
keywords = {conference, important},
focus = {negotiation},
author = {Mohammad, Yasser and Nakadai, Shinji},
title = {Concurrent Negotiations with Global Utility Functions},
booktitle = {Proceedings of the 21st International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2022)},
pages = {1947--1949},
publisher = {IFAAMAS},
address = {Auckland, New Zealand},
year = {2022},
doi = {10.5555/3535850.3536162},
url = {https://www.ifaamas.org/Proceedings/aamas2022/pdfs/p1947.pdf}
}
Automated Negotiation is attracting more attention from researchers recently as it is becoming more relevant to industrial and business applications with increased reliance on automated systems. Most research in this area assumes either a single negotiation thread with a well-defined utility function for each agent involved or a set of concurrent negotiations with an ordering of outcomes in each local negotiation. In this paper, we consider an agent engaged in a set of concurrent negotiations with a utility function defined only for the in of them and no locally defined ordering of outcomes in any negotiation independent from what happens in the others. We argue that this problem setting is interesting both from the academic and the industrial points of view. The paper then presents an algorithm that allows such agent to maximize its expected global utility by orchestrating its behavior in all negotiation threads. The performance of the proposed method is analyzed theoretically and empirically using simulation.
Automated Negotiation
Sengupta, A., Mohammad, Y., & Nakadai, S. (2021). An Autonomous Negotiating Agent Framework with Reinforcement Learning Based Strategies and Adaptive Strategy Switching Mechanism. Proceedings of the 20th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2021), 1163–1172. https://ifaamas.org/Proceedings/aamas2021/pdfs/p1163.pdf
@inproceedings{sengupta2021autonomous,
keywords = {conference, important},
focus = {negotiation},
author = {Sengupta, Ayan and Mohammad, Yasser and Nakadai, Shinji},
title = {An Autonomous Negotiating Agent Framework with Reinforcement Learning Based Strategies and Adaptive Strategy Switching Mechanism},
booktitle = {Proceedings of the 20th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2021)},
pages = {1163--1172},
publisher = {ACM},
year = {2021},
doi = {10.5555/3463952.3464087},
url = {https://ifaamas.org/Proceedings/aamas2021/pdfs/p1163.pdf}
}
Despite abundant negotiation strategies in literature, the complexity of automated negotiation forbids a single strategy from being dominant against all others in different negotiation scenarios. To overcome this, one approach is to use mixture of experts, but at the same time one problem of this method is the selection of experts, as this approach is limited by the competency of the experts selected. Another problem with most negotiation strategies is their incapability of adapting to dynamic variation of the opponent’s behaviour within a single negotiation session resulting in poor performance. This work focuses on both, solving the problem of expert selection and adapting to the opponent’s behaviour with our Autonomous Negotiating Agent Framework. This framework allows real-time classification of opponent’s behaviour and provides a mechanism to select, switch or combine strategies within a single negotiation session. Additionally, our framework has a reviewer component which enables self-enhancement capability by deciding to include new strategies or replace old ones with better strategies periodically. We demonstrate an instance of our framework by implementing maximum entropy reinforcement learning based strategies with a deep learning based opponent classifier. Finally, we evaluate the performance of our agent against state-of-the-art negotiators under varied negotiation scenarios.
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2010). Learning Interaction Protocols Using Augmented Bayesian Networks Applied to Guided Navigation. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2010), 4119–4126.
@inproceedings{mohammad2010augmentedbayes,
keywords = {conference, important},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Learning Interaction Protocols Using Augmented Bayesian Networks Applied to Guided Navigation},
booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2010)},
pages = {4119--4126},
publisher = {IEEE},
address = {Taipei, Taiwan},
year = {2010},
doi = {10.1109/IROS.2010.5651719}
}
Research in robot navigation usually concentrates on implementing navigation algorithms that allow the robot to navigate without human aid. In many real world situations, it is desirable that the robot is able to understand natural gestures from its user or partner and use this understanding to guide its navigation. Some algorithms already exist for learning natural gestures and/or their associated actions but most of these systems does not allow the robot to automatically generate the associated controller that allows it to actually navigate in the real environment. Furthermore, a technique is needed to combine the gestures/actions learned from interacting with multiple users or partners. This paper resolves these two issues and provides a complete system that allows the robot to learn interaction protocols and act upon them using only unsupervised learning techniques and enables it to combine the protocols learned from multiple users/partners. The proposed approach is general and can be applied to other interactive tasks as well. This paper also provides a real world experiment involving 18 subjects and 72 sessions that supports the ability of the proposed system to learn the needed gestures and to improve its knowledge of different gestures and their associations to actions over time.
ML & Time-Series
Mohammad, Y. F. O., & Nishida, T. (2010). Mining Causal Relationships in Multidimensional Time Series. In E. Szczerbicki & N. T. Nguyen (Eds.), Smart Information and Knowledge Management: Advances, Challenges, and Critical Issues (Vol. 260, pp. 309–338). Springer.
@incollection{mohammad2010miningcausal,
keywords = {bookchapter, important},
focus = {timeseries},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Mining Causal Relationships in Multidimensional Time Series},
booktitle = {Smart Information and Knowledge Management: Advances, Challenges, and Critical Issues},
editor = {Szczerbicki, Edward and Nguyen, Ngoc Thanh},
series = {Studies in Computational Intelligence},
volume = {260},
pages = {309--338},
publisher = {Springer},
year = {2010},
doi = {10.1007/978-3-642-04584-4_14}
}
Time series are ubiquitous in all domains of human endeavor. They are generated, stored, and manipulated during any kind of activity. The goal of this chapter is to introduce a novel approach to mine multidimensional time-series data for causal relationships. The main feature of the proposed system is supporting discovery of causal relations based on automatically discovered recurring patterns in the input time series. This is achieved by integrating a variety of data mining techniques. The main insight of the proposed system is that causal relations can be found more easily and robustly by analyzing meaningful events in the time series rather than by analyzing the time series numerical values directly. The RSST (Robust Singular Spectrum Transform) algorithm is used to find interesting points in every time series that is further analyzed by a constrained motif discovery algorithm (if needed) to learn basic events of the time series. The Granger-causality test is extended and applied to the multidimensional time-series describing the occurrences of these basic events rather than to the raw time-series data. The combined algorithm is evaluated using both synthetic and real world data. The real world application is to mine records of activities during a human-robot interaction experiment in which a human subject is guiding a robot to navigate using free hand gesture. The results show that the combined system can provide causality graphs representing the underlying relations between the human’s actions and robot behavior that cannot be recovered using standard causal graph learning procedures. Mining Time Series, Robust Singular Spectrum Transform, Granger-Causality, Mining Causal Relations
ML & Time-Series
Mohammad, Y. F. O., & Nishida, T. (2009). Constrained Motif Discovery in Time Series. New Generation Computing, 27(4), 319–346.
@article{mohammad2009constrainedmotif,
keywords = {journal, important},
focus = {timeseries},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Constrained Motif Discovery in Time Series},
journal = {New Generation Computing},
volume = {27},
number = {4},
pages = {319--346},
year = {2009},
doi = {10.1007/s00354-009-0068-x}
}
The goal of motif discovery algorithms is to efficiently find unknown recurring patterns. In this paper we focus on motif discovery in time series. Most available algorithms cannot utilize domain knowledge in any way which results in quadratic or at least super-linear time and space complexity. In this paper we define the Constrained Motif Discovery problem which enables utilization of domain knowledge into the motif discovery process. The paper then provides two algorithms called MCFull and MCInc for efficiently solving the constrained motif discovery problem. We also show that most unconstrained motif discovery problems be converted into constrained ones using a change-point detection algorithm. A novel change-point detection algorithm called the Robust Singular Spectrum Transform (RSST) is then introduced and compared to traditional Singular Spectrum Transform using synthetic and real-world data sets. The results show that RSST achieves higher specificity and is more adequate for finding constraints to convert unconstrained motif discovery problems to constrained ones that can be solved using MCFull and MCInc. We then compare the combination of RSST and MCFull or MCInc with two state-of-the-art motif discovery algorithms on a large set of synthetic time series. The results show that the proposed algorithms provided four to ten folds increase in speed compared the unconstrained motif discovery algorithms studied without any loss of accuracy. RSST+MCFull is then used in a real world human-robot interaction experiment to enable the robot to learn free hand gestures, actions, and their associations by humans and other robots interacting.
Robotics & HRI
Mohammad, Y. F. O., Nishida, T., & Okada, S. (2009). Unsupervised Simultaneous Learning of Gestures, Actions and Their Associations for Human-Robot Interaction. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2009), 2537–2544.
@inproceedings{mohammad2009simultaneous,
keywords = {conference, important},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki and Okada, Shogo},
title = {Unsupervised Simultaneous Learning of Gestures, Actions and Their Associations for Human-Robot Interaction},
booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2009)},
pages = {2537--2544},
publisher = {IEEE},
address = {St. Louis, MO, USA},
year = {2009},
doi = {10.1109/IROS.2009.5353987}
}
To enable free natural communication between a human operator and a robot three problems must be faced: Firstly the robot have to know the actions it can do in the world. Secondly the robot must be able to learn the patterns in the perceived behavior of its operator that correspond to commands. Finally the robot needs to know when to execute a specific action based on its perception of the operator’s behavior. In this paper we are interested in free hand gestures as the commanding channel. The most restrictive solution to the aforementioned three problems is to fix the action space (pre-programmed actions), fix the command space (predefined gestures), and fix action-command relation (fixed gesture meanings).Learning by demonstration can be viewed as a technique to relax the first restriction by learning the action space. Gesture interpretation can be viewed as a technique to relax the second restriction by learning the command space. Reinforcement learning can be viewed as a technique for relaxing the third restriction by learning action-command associations (policy). In this paper we propose a novel technique that allows the robot to solve these three problems together learning the action space, the command space, and their relations by just another robot operated by a human operator. The main technical contribution of this paper is the introduction of a novel algorithm that allows the robot to segment and discover patterns in its perceived signals without any prior knowledge of the number of different patterns, their occurrences or lengths. The second contribution is using a Ganger-Causality based test to limit the search space for actions and commands utilizing their relations and taking into account the autonomy level of the robot. The paper also presents a feasibility study in which the learning robot was able to predict actor’s behavior with 95.2 after monitoring a single interaction between a novice operator and a WOZ operated robot representing the actor.
Books
Robotics & HRI
Mohammad, Y., & Nishida, T. (2015). Data Mining for Social Robotics: Toward Autonomously Social Robots. Springer.
@book{mohammad2015dataminingbook,
keywords = {book},
focus = {robotics},
author = {Mohammad, Yasser and Nishida, Toyoaki},
title = {Data Mining for Social Robotics: Toward Autonomously Social Robots},
series = {Advanced Information and Knowledge Processing},
publisher = {Springer},
year = {2015},
doi = {10.1007/978-3-319-25232-2},
isbn = {978-3-319-25230-8}
}
Robotics & HRI
Nishida, T., Nakazawa, A., Ohmoto, Y., & Mohammad, Y. (2014). Conversational Informatics: A Data-Intensive Approach with Emphasis on Nonverbal Communication. Springer.
@book{nishida2014conversational,
keywords = {book},
focus = {robotics},
author = {Nishida, Toyoaki and Nakazawa, Atsushi and Ohmoto, Yoshimasa and Mohammad, Yasser},
title = {Conversational Informatics: A Data-Intensive Approach with Emphasis on Nonverbal Communication},
publisher = {Springer},
year = {2014},
isbn = {978-4-431-55039-6}
}
Patents
Automated Negotiation
Mohammad, Y. F. O. (2026). Negotiation Visualization Apparatus, and Negotiation Visualization Method. A method for visualizing automated negotiation processes to enhance usability and a human-machine interface for negotiation support based on that visualization. (100%).
@patent{mohammad2026visualizationpatent,
keywords = {patent},
focus = {negotiation},
author = {Mohammad, Yasser F. O.},
title = {Negotiation Visualization Apparatus, and Negotiation Visualization Method},
number = {JP2026/0000483 (application)},
location = {JP},
note = {A method for visualizing automated negotiation processes to enhance usability and a human-machine interface for negotiation support based on that visualization. (100\%)},
year = {2026}
}
Automated Negotiation
Mohammad, Y. F. O. (2025). Negotiation Protocol, Method and System for Implementing. A system for adapting automated negotiation strategies for the TAU protocol. (100%). https://www.freepatentsonline.com/y2025/0054034.html
@patent{mohammad2025taupatent,
keywords = {patent},
focus = {negotiation},
author = {Mohammad, Yasser F. O.},
title = {Negotiation Protocol, Method and System for Implementing},
number = {US 18/447,264 (application)},
location = {USA},
note = {A system for adapting automated negotiation strategies for the TAU protocol. (100\%)},
url = {https://www.freepatentsonline.com/y2025/0054034.html},
year = {2025}
}
Automated Negotiation
Ando, T., Morinaga, S., & Mohammad, Y. F. O. (2025). Negotiation Device, Negotiation Method, and Storage Medium.
@patent{ando2025devicepatent,
keywords = {patent},
focus = {negotiation},
author = {Ando, Tomohito and Morinaga, Satoshi and Mohammad, Yasser F. O.},
title = {Negotiation Device, Negotiation Method, and Storage Medium},
number = {US 19/095,084 (application)},
location = {USA},
year = {2025}
}
Automated Negotiation
Sengupta, A., Mohammad, Y. F. O., & Nakadai, S. (2024). Automated Negotiation Agent Adaptation. Automated negotiation agent adaptation is performed by detecting change in a utility function. (30%). https://patents.google.com/patent/US12086895B2/en
@patent{sengupta2024adaptationpatent,
keywords = {patent},
focus = {negotiation},
author = {Sengupta, Ayan and Mohammad, Yasser F. O. and Nakadai, Shinji},
title = {Automated Negotiation Agent Adaptation},
number = {US 12,086,895 (granted)},
location = {USA},
note = {Automated negotiation agent adaptation is performed by detecting change in a utility function. (30\%)},
url = {https://patents.google.com/patent/US12086895B2/en},
year = {2024}
}
Automated Negotiation
Mohammad, Y. F. O., & Nakadai, S. (2024). Negotiation Method Including Selection of Neural Network and System for Implementing. A system for selecting and using a neural architecture for automated negotiation. (90%). https://www.freepatentsonline.com/y2024/0386262.html
@patent{mohammad2024nnselectionpatent,
keywords = {patent},
focus = {negotiation},
author = {Mohammad, Yasser F. O. and Nakadai, Shinji},
title = {Negotiation Method Including Selection of Neural Network and System for Implementing},
number = {US 18/317,112 (application)},
location = {USA},
note = {A system for selecting and using a neural architecture for automated negotiation. (90\%)},
url = {https://www.freepatentsonline.com/y2024/0386262.html},
year = {2024}
}
Mohammad, Y. F. O. (2023). Policy Generation Apparatus, Control Method, and Non-Transitory Computer-Readable Storage Medium. A method for concurrent negotiation proven to be optimal against opponents with static acceptance models. (100%). https://patents.google.com/patent/US20230289908A1/en
@patent{mohammad2023policypatent,
keywords = {patent},
author = {Mohammad, Yasser F. O.},
title = {Policy Generation Apparatus, Control Method, and Non-Transitory Computer-Readable Storage Medium},
number = {US 18/005,912 (US20230289908A1); PCT/JP2020/029145},
location = {USA},
note = {A method for concurrent negotiation proven to be optimal against opponents with static acceptance models. (100\%)},
url = {https://patents.google.com/patent/US20230289908A1/en},
year = {2023}
}
Automated Negotiation
Mohammad, Y. F. O., & Ninagawa, K. (2022). Negotiation Method Including Elicitation and System for Implementing. A system for negotiation under uncertainty with elicitation-during-negotiation support using the Value of Information concept. (50%). https://www.freepatentsonline.com/y2022/0366483.html
@patent{mohammad2022elicitationpatent,
keywords = {patent},
focus = {negotiation},
author = {Mohammad, Yasser F. O. and Ninagawa, Kotone},
title = {Negotiation Method Including Elicitation and System for Implementing},
number = {US 17/388,004 (application)},
location = {USA},
note = {A system for negotiation under uncertainty with elicitation-during-negotiation support using the Value of Information concept. (50\%)},
url = {https://www.freepatentsonline.com/y2022/0366483.html},
year = {2022}
}
Automated Negotiation
Sengupta, A., & Mohammad, Y. F. O. (2022). Adaptive Autonomous Negotiation Method and System of Using. A new automated negotiation method that can generalize along negotiation domains and opponents. (30%). https://patents.google.com/patent/US20220108412A1/en
@patent{sengupta2022adaptivepatent,
keywords = {patent},
focus = {negotiation},
author = {Sengupta, Ayan and Mohammad, Yasser F. O.},
title = {Adaptive Autonomous Negotiation Method and System of Using},
number = {US 17/184,590 (application)},
location = {USA},
note = {A new automated negotiation method that can generalize along negotiation domains and opponents. (30\%)},
url = {https://patents.google.com/patent/US20220108412A1/en},
year = {2022}
}
Mohammad, Y. F. O., & Hoashi, K. (2021). Learning Data Generator, Judgment Device and Program. A method for training multiple pipelines of convolutional neural networks to achieve high accuracy in activity recognition. (90%). https://patents.google.com/patent/JP2019087106A/en
@patent{mohammad2021learningdatapatent,
keywords = {patent},
author = {Mohammad, Yasser F. O. and Hoashi, Keiichiro},
title = {Learning Data Generator, Judgment Device and Program},
number = {JP6838259B2 (granted); app. JP2019-087106},
location = {Japan},
note = {A method for training multiple pipelines of convolutional neural networks to achieve high accuracy in activity recognition. (90\%)},
url = {https://patents.google.com/patent/JP2019087106A/en},
year = {2021}
}
Mohammad, Y. F. O., & Hoashi, K. (2020). Neural Network Regulator, Device and Program. A method for compression of neural networks (convolutional or otherwise) based on novel application of feature selection techniques. (90%). https://patents.google.com/patent/JP6838259B2/en
@patent{mohammad2020nnregulatorpatent,
keywords = {patent},
author = {Mohammad, Yasser F. O. and Hoashi, Keiichiro},
title = {Neural Network Regulator, Device and Program},
number = {JP6754343B2 (granted)},
location = {Japan},
note = {A method for compression of neural networks (convolutional or otherwise) based on novel application of feature selection techniques. (90\%)},
url = {https://patents.google.com/patent/JP6838259B2/en},
year = {2020}
}
Invited & Industrial Talks
Mohammad, Y. (2025). Advances in Solving Automated Negotiation Games. Invited talk at the Second Workshop on Game AI Algorithms and Multi-Agent Learning (GAAMAL@IJCAI 2025), Montreal, Canada.
@misc{mohammad2025gaiwtalk,
keywords = {talk},
author = {Mohammad, Yasser},
title = {Advances in Solving Automated Negotiation Games},
howpublished = {Invited talk at the Second Workshop on Game AI Algorithms and Multi-Agent Learning (GAAMAL@IJCAI 2025), Montreal, Canada},
year = {2025}
}
Mohammad, Y. (2025). NEC on Catena-X — Automated Negotiation Technology for the Catena-X Community. Talk at Digital Twin Innovation in Manufacturing, Energy & Related Industries, Heilbronn, Germany.
@misc{mohammad2025catenax,
keywords = {industrialtalk},
author = {Mohammad, Yasser},
title = {{NEC} on {Catena-X} --- Automated Negotiation Technology for the {Catena-X} Community},
howpublished = {Talk at Digital Twin Innovation in Manufacturing, Energy \& Related Industries, Heilbronn, Germany},
year = {2025}
}
Mohammad, Y. (2024). Automated Negotiation: A New Frontier for AI in Business. Invited talk at Integration of Machine Learning and Mathematical Modeling, and Deepening of Its Theory II, Kyushu, Japan.
@misc{mohammad2024kyushutalk,
keywords = {talk},
author = {Mohammad, Yasser},
title = {Automated Negotiation: A New Frontier for {AI} in Business},
howpublished = {Invited talk at Integration of Machine Learning and Mathematical Modeling, and Deepening of Its Theory II, Kyushu, Japan},
year = {2024}
}
Mohammad, Y. (2024). Digital Twin Autonomous Orchestration and Coordination. Talk at the Digital Twin Consortium Q3 Member Meeting, Chicago, IL, USA.
@misc{mohammad2024dtctalk,
keywords = {industrialtalk},
author = {Mohammad, Yasser},
title = {Digital Twin Autonomous Orchestration and Coordination},
howpublished = {Talk at the Digital Twin Consortium Q3 Member Meeting, Chicago, IL, USA},
year = {2024}
}
Mohammad, Y. (2022). Generalized Bargaining Mechanisms: Mechanism Design for Automated Negotiation. Invited talk at the IBM/DIMACS Workshop on Bridging Game Theory and Machine Learning for Multi-Party Decision Making, New Jersey, USA.
@misc{mohammad2022dimacstalk,
keywords = {talk},
author = {Mohammad, Yasser},
title = {Generalized Bargaining Mechanisms: Mechanism Design for Automated Negotiation},
howpublished = {Invited talk at the IBM/DIMACS Workshop on Bridging Game Theory and Machine Learning for Multi-Party Decision Making, New Jersey, USA},
year = {2022}
}
Mohammad, Y. (2022). Concurrent Negotiation in Supply Chains: Problems, Solutions and Challenges. Invited talk at the 13th International Workshop on Agent-Based Complex Automated Negotiations (ACAN@IJCAI 2022), Vienna, Austria.
@misc{mohammad2022acantalk,
keywords = {talk},
author = {Mohammad, Yasser},
title = {Concurrent Negotiation in Supply Chains: Problems, Solutions and Challenges},
howpublished = {Invited talk at the 13th International Workshop on Agent-Based Complex Automated Negotiations (ACAN@IJCAI 2022), Vienna, Austria},
year = {2022}
}
Mohammad, Y. (2018). Analysis and Commentary on PRIANAC and ANAC. Invited talk at the Pacific Rim International Automated Negotiation Agents Competition (PRIANAC@PRIMA 2018), Tokyo, Japan.
@misc{mohammad2018prianactalk,
keywords = {talk},
author = {Mohammad, Yasser},
title = {Analysis and Commentary on {PRIANAC} and {ANAC}},
howpublished = {Invited talk at the Pacific Rim International Automated Negotiation Agents Competition (PRIANAC@PRIMA 2018), Tokyo, Japan},
year = {2018}
}
Mohammad, Y. (2018). Fluid Imitation. Invited talk at Human-Robot Interaction: From Service to Industry (HRI-SI2018@IEEE RO-MAN 2018), Nanjing, China.
@misc{mohammad2018fluidtalk,
keywords = {talk},
author = {Mohammad, Yasser},
title = {Fluid Imitation},
howpublished = {Invited talk at Human-Robot Interaction: From Service to Industry (HRI-SI2018@IEEE RO-MAN 2018), Nanjing, China},
year = {2018}
}
Mohammad, Y. (2012). SSA Application to Motif Discovery and Causality Analysis in Robotics. Invited talk at the Third International Conference on SSA and Its Applications, Beijing, China.
@misc{mohammad2012ssatalk,
keywords = {talk},
author = {Mohammad, Yasser},
title = {{SSA} Application to Motif Discovery and Causality Analysis in Robotics},
howpublished = {Invited talk at the Third International Conference on SSA and Its Applications, Beijing, China},
year = {2012}
}
Tutorials
Mohammad, Y. (2025). Developing Data-Driven Automated Negotiating Agents. Tutorial at the 29th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD 2025), Sydney, Australia.
@misc{mohammad2025pakddtutorial,
keywords = {tutorial},
author = {Mohammad, Yasser},
title = {Developing Data-Driven Automated Negotiating Agents},
howpublished = {Tutorial at the 29th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD 2025), Sydney, Australia},
year = {2025}
}
Mohammad, Y. (2025). Reinforcement Learning for Automated Negotiation. Tutorial at the 24th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2025), Detroit, MI, USA.
@misc{mohammad2025aamastutorial,
keywords = {tutorial},
author = {Mohammad, Yasser},
title = {Reinforcement Learning for Automated Negotiation},
howpublished = {Tutorial at the 24th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2025), Detroit, MI, USA},
year = {2025}
}
Mohammad, Y. (2023). Automated Negotiation in Supply Chains. Tutorial at the IEEE International Conference on Agents (ICA 2023), Kyoto, Japan.
@misc{mohammad2023icatutorial,
keywords = {tutorial},
author = {Mohammad, Yasser},
title = {Automated Negotiation in Supply Chains},
howpublished = {Tutorial at the IEEE International Conference on Agents (ICA 2023), Kyoto, Japan},
year = {2023}
}
Mohammad, Y. (2023). Reinforcement Learning for Automated Negotiation. Tutorial at the Australasian Joint Conference on Artificial Intelligence (AJCAI 2023), Brisbane, Australia.
@misc{mohammad2023ajcaitutorial,
keywords = {tutorial},
author = {Mohammad, Yasser},
title = {Reinforcement Learning for Automated Negotiation},
howpublished = {Tutorial at the Australasian Joint Conference on Artificial Intelligence (AJCAI 2023), Brisbane, Australia},
year = {2023}
}
Mohammad, Y., & Greenwald, A. (2022). Automated Negotiation: Challenges and Tools. Tutorial at the 36th AAAI Conference on Artificial Intelligence (AAAI 2022), Vancouver, Canada.
@misc{mohammad2022aaaitutorial,
keywords = {tutorial},
author = {Mohammad, Yasser and Greenwald, Amy},
title = {Automated Negotiation: Challenges and Tools},
howpublished = {Tutorial at the 36th AAAI Conference on Artificial Intelligence (AAAI 2022), Vancouver, Canada},
year = {2022}
}
Mohammad, Y. (2020). Automated Negotiation in Supply Chain Management. Tutorial at the 23rd International Conference on Principles and Practice of Multi-Agent Systems (PRIMA 2020), Nagoya, Japan.
@misc{mohammad2020primatutorial,
keywords = {tutorial},
author = {Mohammad, Yasser},
title = {Automated Negotiation in Supply Chain Management},
howpublished = {Tutorial at the 23rd International Conference on Principles and Practice of Multi-Agent Systems (PRIMA 2020), Nagoya, Japan},
year = {2020}
}
Mohammad, Y. (2019). Automated Negotiation: Challenges and Tools. Tutorial at the 22nd International Conference on Principles and Practice of Multi-Agent Systems (PRIMA 2019), Turin, Italy.
@misc{mohammad2019primatutorial,
keywords = {tutorial},
author = {Mohammad, Yasser},
title = {Automated Negotiation: Challenges and Tools},
howpublished = {Tutorial at the 22nd International Conference on Principles and Practice of Multi-Agent Systems (PRIMA 2019), Turin, Italy},
year = {2019}
}
Journal Articles
Automated Negotiation
Mohammad, Y. (2026). Automated Negotiation with No Information about Partner Utility Functions Using the Tentative Acceptance Unique Offers Protocol. Autonomous Agents and Multi-Agent Systems, 40(1), 21.
@article{mohammad2026tau,
keywords = {journal, important},
focus = {negotiation},
author = {Mohammad, Yasser},
title = {Automated Negotiation with No Information about Partner Utility Functions Using the Tentative Acceptance Unique Offers Protocol},
journal = {Autonomous Agents and Multi-Agent Systems},
volume = {40},
number = {1},
pages = {21},
year = {2026},
doi = {10.1007/s10458-026-09745-9}
}
With the widespread adoption of AI in industrial and business operations, finding methods to reach agreement between intelligent agents representing self-interested entities in a general-sum environment (cooperation within competition) is attracting more interest in the research community. Negotiation is a common process for reaching agreements between people and human institutions. Automated negotiation is thus being considered for cooperation within competition situations involving AIs. The most widely used protocols for automated negotiation are the Stacked Alternating Offers Protocol (SAOP) for multilateral negotiations and the Alternating Offers Protocol (AOP) for bilateral negotiations which directly model bargaining as in human marketplaces. Several strategies have been proposed for these protocols over the years. In this paper, we propose a modification of the AOP and SAOP protocols and a method for adapting negotiation strategies to the new protocol. We show empirically that the proposed approach leads to higher expected advantage for all agents, and achieves higher agreement rate, higher welfare, and fairer agreements faster. This is achieved at the expense of a small increase in information revelation.
Automated Negotiation
Mohammad, Y., Chen, H., Higa, R., Ando, T., & Morinaga, S. (2025). Generative AI for Automated Negotiation. Journal of Innovation.
@article{mohammad2025generativeai,
keywords = {journal},
focus = {negotiation},
author = {Mohammad, Yasser and Chen, Haifeng and Higa, Ryota and Ando, Tomohito and Morinaga, Satoshi},
title = {Generative {AI} for Automated Negotiation},
journal = {Journal of Innovation},
year = {2025},
month = dec
}
ML & Time-Series
Petch, L., Moustafa, A., Ma, X., & Mohammad, Y. (2025). HFL-GAN: Scalable Hierarchical Federated Learning GAN for High Quantity Heterogeneous Clients. Applied Intelligence, 55, 170.
@article{petch2025hflgan,
keywords = {journal},
focus = {timeseries},
author = {Petch, Lewis and Moustafa, Ahmed and Ma, Xinhui and Mohammad, Yasser},
title = {{HFL-GAN}: Scalable Hierarchical Federated Learning {GAN} for High Quantity Heterogeneous Clients},
journal = {Applied Intelligence},
volume = {55},
pages = {170},
year = {2025},
doi = {10.1007/s10489-024-05924-x}
}
Hamdi, F. A., Kataoka, K., Arai, Y., Takeda, N., Yamamoto, M., Mohammad, Y. F. O., Ghazy, N. A., & Suzuki, T. (2023). An Octopamine Receptor Involved in Feeding Behavior of the Two-Spotted Spider Mite, Tetranychus urticae Koch: A Possible Candidate for RNAi-Based Pest Control. Entomologia Generalis, 43(1).
@article{hamdi2023octopamine,
keywords = {journal},
author = {Hamdi, Faten Abdelsalam and Kataoka, Kosuke and Arai, Yuka and Takeda, Naoki and Yamamoto, Masanobu and Mohammad, Yasser F. O. and Ghazy, Noureldin Abuelfadl and Suzuki, Takeshi},
title = {An Octopamine Receptor Involved in Feeding Behavior of the Two-Spotted Spider Mite, Tetranychus urticae {Koch}: A Possible Candidate for {RNAi}-Based Pest Control},
journal = {Entomologia Generalis},
volume = {43},
number = {1},
year = {2023}
}
Automated Negotiation
Mohammad, Y. (2023). Optimal Time-Based Strategy for Automated Negotiation. Applied Intelligence, 53(6), 6710–6735.
@article{mohammad2023optimaltime,
keywords = {journal, important},
focus = {negotiation},
author = {Mohammad, Yasser},
title = {Optimal Time-Based Strategy for Automated Negotiation},
journal = {Applied Intelligence},
volume = {53},
number = {6},
pages = {6710--6735},
year = {2023},
doi = {10.1007/s10489-022-03662-6}
}
Recent years are showing increased adoption of AI technology to automate business and production processes thanks to the recent successes of machine learning techniques. This leads to increased interest in automated negotiation as a method for achieving win-win agreements among self-interested agents. Research in automated negotiation can be traced back to the Nash bargaining game in the mid 20 th century. Nevertheless, finding an optimal negotiation strategy against an unknown opponent with an unknown utility function is still an open area of research. The most recent result in this area is the Greedy Concession Algorithm (GCA) which can be shown to be optimal under specific constraints on both the negotiation protocol (non-repeating offers), opponent (static acceptance-model) and search space (deterministic time-based strategies). In this paper, we extend this line of work by providing an algorithmically faster version of GCA called Quick GCA which reduces the time-complexity of the search process from O( 2 ) to O( ) where is the size of the outcome-space and is the number of negotiation rounds allowed. Moreover, we show that GCA/QGCA can be applied in a more general setting; Namely with repeating-offers protocols and to search the more general probabilistic time-based strategies. Finally, we heuristically extend QGCA to more general opponents with general time-dependent acceptance-model and negotiation settings (real-time limited negotiations) in three steps called , , and that iteratively and greedily modify the policy proposed by applied to an approximate static acceptance model . The paper evaluates the proposed approach empirically against state of the art negotiation strategies (winners of all relevant ANAC competition winners) and shows that it outperforms them in a wide variety of negotiation scenarios.
Automated Negotiation
Mohammad, Y. (2021). Concurrent Local Negotiations with a Global Utility Function: A Greedy Approach. Autonomous Agents and Multi-Agent Systems, 35(2), 28.
@article{mohammad2021concurrentlocal,
keywords = {journal},
focus = {negotiation},
author = {Mohammad, Yasser},
title = {Concurrent Local Negotiations with a Global Utility Function: A Greedy Approach},
journal = {Autonomous Agents and Multi-Agent Systems},
volume = {35},
number = {2},
pages = {28},
year = {2021},
doi = {10.1007/s10458-021-09512-y}
}
Automated Negotiation is a growing area of research in recent years as it provides a mechanism for intelligent agents representing people and institutions to coordinate their behavior in a complex environment under rational selfish assumptions. Most research in this area assumes either a single negotiation thread with a well-defined utility function for each agent involved or a set of concurrent negotiations with an ordering of outcomes in each local negotiation. In this paper, we consider an agent engaging in a set of concurrent negotiations with the utility function only defined for the in of them and no locally defined ordering of outcomes in any negotiation. The paper presents an algorithm that allows such agent to maximize its expected global utility function by orchestrating its behavior in all negotiation threads. The performance of the proposed method is analyzed theoretically and empirically using simulation in the context of a trading market.
Automated Negotiation
Mohammad, Y., Nakadai, S., Morinaga, S., & Fujita, K. (2020). Supply Chain Management League (SCML) — Automated Negotiating Agent Competition for Manufacturing Value Chain. Journal of the Japanese Society for Artificial Intelligence, 35(3).
@article{mohammad2020scmljournal,
keywords = {journal},
focus = {negotiation},
author = {Mohammad, Yasser and Nakadai, Shinji and Morinaga, Satoshi and Fujita, Katsuhide},
title = {Supply Chain Management League {(SCML)} --- Automated Negotiating Agent Competition for Manufacturing Value Chain},
journal = {Journal of the Japanese Society for Artificial Intelligence},
volume = {35},
number = {3},
year = {2020},
month = may
}
ML & Time-Series
Mohammad, Y. F. O., Matsumoto, K., & Hoashi, K. (2019). Selecting Orientation-Insensitive Features for Activity Recognition from Accelerometers. IEICE Transactions on Information and Systems, E102-D(1), 104–115.
@article{mohammad2019selecting,
keywords = {journal},
focus = {timeseries},
author = {Mohammad, Yasser F. O. and Matsumoto, Kazunori and Hoashi, Keiichiro},
title = {Selecting Orientation-Insensitive Features for Activity Recognition from Accelerometers},
journal = {IEICE Transactions on Information and Systems},
volume = {E102-D},
number = {1},
pages = {104--115},
year = {2019},
doi = {10.1587/transinf.2018EDP7092}
}
ML & Time-Series
Mohammad, Y. F. O., Matsumoto, K., & Hoashi, K. (2018). Primitive Activity Recognition from Short Sequences of Sensory Data. Applied Intelligence, 48(10), 3748–3761.
@article{mohammad2018primitive,
keywords = {journal},
focus = {timeseries},
author = {Mohammad, Yasser F. O. and Matsumoto, Kazunori and Hoashi, Keiichiro},
title = {Primitive Activity Recognition from Short Sequences of Sensory Data},
journal = {Applied Intelligence},
volume = {48},
number = {10},
pages = {3748--3761},
year = {2018},
doi = {10.1007/s10489-018-1166-6}
}
Activity recognition (AR) from mobile device sensors and wearables is attracting more attention from the research community due to the widespread adoption of these devices and the unique opportunity they provide for understanding user’s behavior leading to novel services and improvements in the delivery of existing ones. Approaches to tackle this problem either rely on predefined statistical features of sensor data streams or feature learning with the latter providing higher accuracies in most cases. Deep learning methods proved more effective than traditional approaches to feature learning in multiple studies. This paper presents a novel end-to-end trainable deep architecture that utilizes multiple convolutional neural networks (CNN), late fusion and extensive layer bypassing. The proposed method can easily accommodate multiple sensors and signal representations. The proposed approach is validated on eight publicly available datasets using a variety of evaluation conditions showing that it outperforms state-of-the-art methods in six of them.
ML & Time-Series
Mohammad, Y. F. O., & Nishida, T. (2016). Exact Multi-Length Scale and Mean Invariant Motif Discovery. Applied Intelligence, 44(2), 322–339.
@article{mohammad2016exactmulti,
keywords = {journal},
focus = {timeseries},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Exact Multi-Length Scale and Mean Invariant Motif Discovery},
journal = {Applied Intelligence},
volume = {44},
number = {2},
pages = {322--339},
year = {2016},
doi = {10.1007/s10489-015-0684-8}
}
Discovering approximately recurrent motifs (ARMs) in timeseries is an active area of research in data mining. Exact motif discovery is defined as the problem of efficiently finding the most similar pairs of timeseries subsequences and can be used as a basis for discovering ARMs. The most efficient algorithm for solving this problem was the MK algorithm which was designed to find a single pair of timeseries subsequences with maximum similarity at a known length. This paper provides three of extensions of the MK algorithm that allow it to find the top K similar subsequences at multiple lengths using both the Euclidean distance metric and scale invariant normalized version of it. The proposed algorithms are then applied to both synthetic data and real-world data with a focus on discovery of ARMs in human motion trajectories. Data Mining Motif Discovery HRI Human Behavior Understanding Exact Motif Discovery
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2015). Why Should We Imitate Robots? Effect of Back Imitation on Judgment of Imitative Skill. International Journal of Social Robotics, 7(4), 497–512.
@article{mohammad2015whyimitate,
keywords = {journal},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Why Should We Imitate Robots? {Effect} of Back Imitation on Judgment of Imitative Skill},
journal = {International Journal of Social Robotics},
volume = {7},
number = {4},
pages = {497--512},
year = {2015},
doi = {10.1007/s12369-015-0282-2}
}
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2015). Learning Interaction Protocols by Mimicking: Understanding and Reproducing Human Interactive Behavior. Pattern Recognition Letters, 66, 62–70.
@article{mohammad2015mimicking,
keywords = {journal},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Learning Interaction Protocols by Mimicking: Understanding and Reproducing Human Interactive Behavior},
journal = {Pattern Recognition Letters},
volume = {66},
pages = {62--70},
year = {2015},
doi = {10.1016/j.patrec.2014.11.010}
}
ML & Time-Series
Mohammad, Y. F. O., & Nishida, T. (2015). Shift Density Estimation Based Approximately Recurring Motif Discovery. Applied Intelligence, 42(1), 112–134.
@article{mohammad2015shiftdensity,
keywords = {journal},
focus = {timeseries},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Shift Density Estimation Based Approximately Recurring Motif Discovery},
journal = {Applied Intelligence},
volume = {42},
number = {1},
pages = {112--134},
year = {2015},
doi = {10.1007/s10489-014-0531-3}
}
Approximately Recurring Motif (ARM) discovery is the problem of finding unknown patterns that appear frequently in real valued timeseries. In this paper, we propose a novel algorithm for solving this problem that can achieve performance comparable with the most accurate algorithms with a speed comparable to the fastest ones. The main idea behind the proposed algorithm is to convert the problem of ARM discovery into a density estimation problem in the single dimensionality shift-space (rather than in the original time-series space). This makes the algorithm more robust to short noise bursts that can dramatically affect the performance of most available algorithms. The paper also reports the results of applying the proposed algorithm to synthetic and three real-world datasets in the domains of gesture discovery and motion primitive discovery. Data Mining Motif Discovery HRI
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2013). Learning Where to Look: Autonomous Development of Gaze Behavior for Natural Human-Robot Interaction. Interaction Studies, 14(3), 419–450.
@article{mohammad2013learningwhere,
keywords = {journal},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Learning Where to Look: Autonomous Development of Gaze Behavior for Natural Human-Robot Interaction},
journal = {Interaction Studies},
volume = {14},
number = {3},
pages = {419--450},
year = {2013}
}
Gaze is one of the most important nonverbal behaviors in regulating human-human interactions in close encounters. Several approaches have been proposed to design robots that can exhibit human-like natural gazing behavior. Most of the available techniques though are based on careful hard-coding of behavioral rules that are usually extracted from analysis of human-human interaction corpora or theories of gaze behavior in humans. The major disadvantage of this approach is the difficulty in discovering these behavioral rules specially taking into account cultural effects, and the effect of power distribution on gaze behavior. In this paper, we report the development and evaluation of an unsupervised gaze controller for a listener robot that tries to alleviate this problem of engineered gazing behavior. The system utilizes two major learning mechanisms. First, it learns a set of recurrent gaze patterns called basic interactive acts using a motif discovery algorithm. A hierarchical controller to activate these acts as needed during the interaction is then learned resulting in a grounded controller. The system was implemented and evaluated in comparison with a reactive gaze controller that was previously shown to provide human-like gazing behavior. The proposed system was shown to be superior in terms of naturalness, human-likeness and comfort of the human partner.
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2012). Fluid Imitation: Discovering What to Imitate. International Journal of Social Robotics, 4(4), 369–382.
@article{mohammad2012fluidjournal,
keywords = {journal},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Fluid Imitation: Discovering What to Imitate},
journal = {International Journal of Social Robotics},
volume = {4},
number = {4},
pages = {369--382},
year = {2012},
doi = {10.1007/s12369-012-0153-z}
}
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2010). Using Physiological Signals to Detect Natural Interactive Behavior. Applied Intelligence, 33(1), 79–92.
@article{mohammad2010physiological,
keywords = {journal},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Using Physiological Signals to Detect Natural Interactive Behavior},
journal = {Applied Intelligence},
volume = {33},
number = {1},
pages = {79--92},
year = {2010},
doi = {10.1007/s10489-010-0241-4}
}
Many researchers in the HRI and ECA domains try to build robots and agents that exhibit human-like behavior in real-world close encounter situations. One major requirement for comparing such robots and agents is to have an objective quantitative metric for measuring naturalness in various kinds of interactions. Some researchers have already suggested techniques for measuring stress level, awareness etc using physiological signals like GSR and BVP. One problem of available techniques is that they are only tested with extreme situations and cannot according to the analysis provided in this paper distinguish the response of human subjects in natural interaction situations. One other problem of the available techniques is that most of them require calibration and some times ad-hoc adjustment for every subject. This paper explores the usefulness of various kinds of physiological signals and statistics in distinguishing natural and unnatural partner behavior in a close encounter situation. The paper also explores the usefulness of these statistics in various time slots of the interaction. Based on this analysis a regressor was designed to measure naturalness in close encounter situations and was evaluated using human-human and human-robot interactions and shown to achieve 87.5 situations. RSST Psychophysiology HRI data mining
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2010). Controlling Gaze with an Embodied Interactive Control Architecture. Applied Intelligence, 32(2), 148–163.
@article{mohammad2010controllinggaze,
keywords = {journal},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Controlling Gaze with an Embodied Interactive Control Architecture},
journal = {Applied Intelligence},
volume = {32},
number = {2},
pages = {148--163},
year = {2010},
doi = {10.1007/s10489-009-0180-0}
}
Human-Robot Interaction (HRI) is a growing field of research that targets the development of robots which are easy to operate, more engaging and more entertaining. Natural human-like behavior is considered by many researchers as an important target of HRI. Research in Human-Human communications revealed that gaze control is one of the major interactive behaviors used by humans in close encounters. Human-like gaze control is then one of the important behaviors that a robot should have in order to provide natural interactions with human partners. To develop human-like natural gaze control that can integrate easily with other behaviors of the robot, a flexible robotic architecture is needed. Most robotic architectures available were developed with autonomous robots in mind. Although robots developed for HRI are usually autonomous, their autonomy is combined with interactivity, which adds more challenges on the design of the robotic architectures supporting them. This paper reports the development and evaluation of two gaze controllers using a new cross-platform robotic architecture for HRI applications called EICA (The Embodied Interactive Control Architecture), that was designed to meet those challenges emphasizing how low level attention focusing and action integration are implemented. Evaluation of the gaze controllers revealed human-like behavior in terms of mutual attention, gaze toward partner, and mutual gaze. The paper also reports a novel Floating Point Genetic Algorithm (FPGA) for learning the parameters of various processes of the gaze controller. Robotic Architectures Action Integration HRI Gaze Control
ML & Time-Series
Mohammad, Y. F. O., & Nishida, T. (2009). Constrained Motif Discovery in Time Series. New Generation Computing, 27(4), 319–346.
@article{mohammad2009constrainedmotif,
keywords = {journal, important},
focus = {timeseries},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Constrained Motif Discovery in Time Series},
journal = {New Generation Computing},
volume = {27},
number = {4},
pages = {319--346},
year = {2009},
doi = {10.1007/s00354-009-0068-x}
}
The goal of motif discovery algorithms is to efficiently find unknown recurring patterns. In this paper we focus on motif discovery in time series. Most available algorithms cannot utilize domain knowledge in any way which results in quadratic or at least super-linear time and space complexity. In this paper we define the Constrained Motif Discovery problem which enables utilization of domain knowledge into the motif discovery process. The paper then provides two algorithms called MCFull and MCInc for efficiently solving the constrained motif discovery problem. We also show that most unconstrained motif discovery problems be converted into constrained ones using a change-point detection algorithm. A novel change-point detection algorithm called the Robust Singular Spectrum Transform (RSST) is then introduced and compared to traditional Singular Spectrum Transform using synthetic and real-world data sets. The results show that RSST achieves higher specificity and is more adequate for finding constraints to convert unconstrained motif discovery problems to constrained ones that can be solved using MCFull and MCInc. We then compare the combination of RSST and MCFull or MCInc with two state-of-the-art motif discovery algorithms on a large set of synthetic time series. The results show that the proposed algorithms provided four to ten folds increase in speed compared the unconstrained motif discovery algorithms studied without any loss of accuracy. RSST+MCFull is then used in a real world human-robot interaction experiment to enable the robot to learn free hand gestures, actions, and their associations by humans and other robots interacting.
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2009). Toward Combining Autonomy and Interactivity for Social Robots. AI & Society, 24(1), 35–49.
@article{mohammad2009combining,
keywords = {journal},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Toward Combining Autonomy and Interactivity for Social Robots},
journal = {AI \& Society},
volume = {24},
number = {1},
pages = {35--49},
year = {2009},
doi = {10.1007/s00146-009-0196-3}
}
Mohammad, Y. F. O., & Nishida, T. (2009). Interactive Perception for Amplification of Intended Behavior in Complex Noisy Environments. AI & Society, 23(2), 167–186.
@article{mohammad2009interactiveperception,
keywords = {journal},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Interactive Perception for Amplification of Intended Behavior in Complex Noisy Environments},
journal = {AI \& Society},
volume = {23},
number = {2},
pages = {167--186},
year = {2009},
doi = {10.1007/s00146-007-0137-y}
}
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2008). Interaction Between Untrained Users and a Miniature Robot in a Collaborative Navigation Controlled Experiment. International Journal of Information Acquisition, 5(4), 291–307.
@article{mohammad2008untrained,
keywords = {journal},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Interaction Between Untrained Users and a Miniature Robot in a Collaborative Navigation Controlled Experiment},
journal = {International Journal of Information Acquisition},
volume = {5},
number = {4},
pages = {291--307},
year = {2008},
doi = {10.1142/S0219878908001727}
}
Future robots are expected to be operated by untrained persons using natural means of communication. This entails the necessity of understanding how humans will communicate with such robots especially in the non humanoid case where anthropomorphism is in its minimum. This paper presents a controlled experiment to study the interaction between untrained human users and a miniature robot in a collaborative navigation task. Three dimensions of the interaction are studied: 1. How the human operators tended to use gestures during their interaction. Interesting results about the patterns of gesture use are reported. 2. Signs of human adaptation to the task requirements and robot capabilities. Three main findings about the human adaptation are reported. 3. The effectiveness and naturalness of using motion cues as a feedback mechanism from the robot in comparison with verbal feedback. The results of the experiment showed that there is no significant difference in the task completion accuracy and time or in the feeling of naturalness between motion cues and verbal feedback, and there is a statistically significant improvement when using either of them compared with the control case. Moreover the subjects selected the motion cues feedback mechanism more frequently as the preferred feedback modality for them.
Robotics & HRI
Nishida, T., Terada, K., Tajima, T., Hatakeyama, M., Ogasawara, Y., Sumi, Y., Xu, Y., Mohammad, Y. F. O., Tarasenko, K., Ohya, T., & Hiramatsu, T. (2006). Toward Robots as Embodied Knowledge Media. IEICE Transactions on Information and Systems, E89-D(6), 1768–1780.
@article{nishida2006embodiedknowledge,
keywords = {journal},
focus = {robotics},
author = {Nishida, Toyoaki and Terada, Kazunori and Tajima, Takashi and Hatakeyama, Makoto and Ogasawara, Yoshiyasu and Sumi, Yasuyuki and Xu, Yong and Mohammad, Yasser F. O. and Tarasenko, Kateryna and Ohya, Taku and Hiramatsu, Tatsuya},
title = {Toward Robots as Embodied Knowledge Media},
journal = {IEICE Transactions on Information and Systems},
volume = {E89-D},
number = {6},
pages = {1768--1780},
year = {2006},
doi = {10.1093/ietisy/e89-d.6.1768}
}
Refereed Book Chapters
Robotics & HRI
Mohammad, Y. (2018). Natural Human-Robot Interaction. In K. L. Norman & J. Kirakowski (Eds.), The Wiley Handbook of Human Computer Interaction (Vol. 2, pp. 641–655). Wiley.
@incollection{mohammad2018naturalhri,
keywords = {bookchapter},
focus = {robotics},
author = {Mohammad, Yasser},
title = {Natural Human-Robot Interaction},
booktitle = {The Wiley Handbook of Human Computer Interaction},
editor = {Norman, Kent L. and Kirakowski, Jurek},
volume = {2},
pages = {641--655},
publisher = {Wiley},
year = {2018}
}
Nonverbal behavior during human-human close encounters is critical to the accomplishment of natural interaction. For this reason, humanoid robots trying to achieve natural interactions with humans should be able to understand and synthesis nonverbal behavior in a way that mimics the human use of it. One of the most important situations during natural human-robot interactions is the explanation scenario in which the human is explaining a task to the robot using natural verbal and nonverbal behavior. This situation occurs frequently in many HRI applications and is critical to the success of the Robots as Knowledge Media project suggested by the authors. In this paper the implementation of a humanoid robot that can show human like gaze control during explanation settings based only on reactive processing is presented. The software of the robot is based on the EICA architecture designed to combine autonomy with interactivity in the lowest level of the system. The details of the implementation and analysis of the naturalness of behavior and the effect of noisy input is presented in this paper.
Robotics & HRI
Mohammad, Y., & Nishida, T. (2010). Modelling Interaction Dynamics During Face-to-Face Interactions. In T. Nishida, L. C. Jain, & C. Faucher (Eds.), Modeling Machine Emotions for Realizing Intelligence — Foundations and Applications (Vol. 1, pp. 53–87). Springer.
@incollection{mohammad2010interactiondynamics,
keywords = {bookchapter},
focus = {robotics},
author = {Mohammad, Yasser and Nishida, Toyoaki},
title = {Modelling Interaction Dynamics During Face-to-Face Interactions},
booktitle = {Modeling Machine Emotions for Realizing Intelligence --- Foundations and Applications},
editor = {Nishida, Toyoaki and Jain, Lakhmi C. and Faucher, Colette},
series = {Smart Innovation, Systems and Technologies},
volume = {1},
pages = {53--87},
publisher = {Springer},
year = {2010},
doi = {10.1007/978-3-642-12604-8_4}
}
During face to face interactions, the emotional state of each participant is greatly affected by the behavior of other participants and how much this behavior conforms with common protocols of interaction in the society. Research in human to human interaction in face to face situations has uncovered many forms of synchrony in the behavior of the interacting partners. This includes factors as body alignment, entrainment of verbal behavior. Maintenance of these kinds of synchrony is essential to keep the interaction natural and to regulate the affective state of the interacting partners. In this chapter we examine the interplay between one partner’s use of interaction protocols, maintenance of synchrony and the emotional response of the other partner in the two way interactions. We will first define the notion of interaction protocol and relate it with the Reactive Theory of Intention and Low Level Emotions. We will then show empirically that the use of suitable interaction protocols is essential to maintain a positive emotional response of the interaction partner during face to face explanation situations. The analysis in this section is based on the H 3R interaction corpus containing sixty six human-human and human-robot interaction sessions. This interaction corpus utilizes physiological, behavioral and subjective data. Using this result, it is necessary to model not only the affective state of the interacting partners but also the interaction protocol that each of them is using. Human-Robot interaction experiments can be of value in analyzing the interaction protocols used by the partners and modelling their emotional response to these protocols. We used Human-Robot interactions in explanation and collaborative navigation tasks as a test-bed for our analysis of interaction protocol emergence and adaptation. The first experiment analyzes how the requirement to maintain the interaction protocol and synchrony restricts the design of the robot and how did we meet these restriction in a semi-autonomous miniature robot. We focus on how low level emotions can be used to act as a mediator between Perception and Behavior. The second experiment explores a computational model of the interaction protocol and evaluates it in an explanation face to face scenario. The chapter also provides a critical analysis of the interplay between interaction protocols and the emotional state of interaction partners. Interaction Dynamics, Embodied Interactive Control Architecture
ML & Time-Series
Mohammad, Y. F. O., & Nishida, T. (2010). Mining Causal Relationships in Multidimensional Time Series. In E. Szczerbicki & N. T. Nguyen (Eds.), Smart Information and Knowledge Management: Advances, Challenges, and Critical Issues (Vol. 260, pp. 309–338). Springer.
@incollection{mohammad2010miningcausal,
keywords = {bookchapter, important},
focus = {timeseries},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Mining Causal Relationships in Multidimensional Time Series},
booktitle = {Smart Information and Knowledge Management: Advances, Challenges, and Critical Issues},
editor = {Szczerbicki, Edward and Nguyen, Ngoc Thanh},
series = {Studies in Computational Intelligence},
volume = {260},
pages = {309--338},
publisher = {Springer},
year = {2010},
doi = {10.1007/978-3-642-04584-4_14}
}
Time series are ubiquitous in all domains of human endeavor. They are generated, stored, and manipulated during any kind of activity. The goal of this chapter is to introduce a novel approach to mine multidimensional time-series data for causal relationships. The main feature of the proposed system is supporting discovery of causal relations based on automatically discovered recurring patterns in the input time series. This is achieved by integrating a variety of data mining techniques. The main insight of the proposed system is that causal relations can be found more easily and robustly by analyzing meaningful events in the time series rather than by analyzing the time series numerical values directly. The RSST (Robust Singular Spectrum Transform) algorithm is used to find interesting points in every time series that is further analyzed by a constrained motif discovery algorithm (if needed) to learn basic events of the time series. The Granger-causality test is extended and applied to the multidimensional time-series describing the occurrences of these basic events rather than to the raw time-series data. The combined algorithm is evaluated using both synthetic and real world data. The real world application is to mine records of activities during a human-robot interaction experiment in which a human subject is guiding a robot to navigate using free hand gesture. The results show that the combined system can provide causality graphs representing the underlying relations between the human’s actions and robot behavior that cannot be recovered using standard causal graph learning procedures. Mining Time Series, Robust Singular Spectrum Transform, Granger-Causality, Mining Causal Relations
Mohammad, Y., & Nishida, T. (2009). Learning Interaction Structure Using a Hierarchy of Dynamical Systems. In B.-C. Chien & T.-P. Hong (Eds.), Opportunities and Challenges for Next-Generation Applied Intelligence (Vol. 214, pp. 253–258). Springer.
@incollection{mohammad2009hierarchy,
keywords = {bookchapter},
author = {Mohammad, Yasser and Nishida, Toyoaki},
title = {Learning Interaction Structure Using a Hierarchy of Dynamical Systems},
booktitle = {Opportunities and Challenges for Next-Generation Applied Intelligence},
editor = {Chien, Been-Chian and Hong, Tzung-Pei},
series = {Studies in Computational Intelligence},
volume = {214},
pages = {253--258},
publisher = {Springer},
year = {2009}
}
The IAM (Interaction Adaptation Manager) algorithm was recently proposed to learn the optimal parameters of a hierarchical dynamical system incrementally through interacting with other agents given that the structure of the system is known (the number of processes in each layer and their interconnections) and that the agent knows how to interact in all roles except the one it is learning (e.g. an agent learning to listen should know how to speak). This paper presents an algorithm for learning the structure of a hierarchical dynamical system representing the interaction protocol at various timescales and using multiple modalities relaxing these two constraint. The proposed system was tested in a simulation environment in which rich human-like agents are interacting and showed accurate recognition of the interaction structure using few training examples. The learned structure showed acceptable performance that allowed subsequent application of the adaptation algorithm to converge to a good solution using as few as 15 interactions. The paper also presents an experiment to evaluate the real-world behavior of a gaze controller learned by the system. The results show that the proposed algorithm outperforms another state-of-the-art gaze controller in terms of human-likeness, apparent understanding of the robot, and comfort of the human partner.
International Conferences
ML & Time-Series
Mohammad, Y., & Chen, H. (2026). Automated Negotiation and Multimodal Time-Series Forecasting for Efficient Procurement. Proceedings of the 25th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2026).
@inproceedings{mohammad2026forecasting,
keywords = {conference},
focus = {timeseries},
author = {Mohammad, Yasser and Chen, Haifeng},
title = {Automated Negotiation and Multimodal Time-Series Forecasting for Efficient Procurement},
booktitle = {Proceedings of the 25th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2026)},
address = {Paphos, Cyprus},
year = {2026}
}
Procurement is a key function in supply chain management that involves acquiring goods and services to meet organizational needs. Efficient procurement is crucial for minimizing costs, ensuring timely delivery, and maintaining quality standards. This paper explores the integration of automated negotiation and multimodal time-series forecasting to enhance procurement processes. Automated negotiation can streamline interactions with suppliers, while multimodal time-series forecasting can improve demand prediction accuracy by leveraging diverse data sources leading to better negotiation outputs. By combining these approaches, organizations can optimize procurement strategies, reduce costs, and improve overall supply chain efficiency. We present two case studies using simulations based on real-world data for procurement that show the effectiveness of the proposed framework.
Automated Negotiation
Aydoğan, R., Baarslag, T., Florijn, T. C. P., Fujita, K., Jonker, C. M., & Mohammad, Y. (2026). [COMP25] The Automated Negotiating Agents Competition (ANAC) 2025 Challenges and Results. CoRR, abs/2604.13914. https://arxiv.org/abs/2604.13914
@inproceedings{aydogan2026anac2025,
keywords = {conference},
focus = {negotiation},
author = {Aydo{\u{g}}an, Reyhan and Baarslag, Tim and Florijn, Tamara C. P. and Fujita, Katsuhide and Jonker, Catholijn M. and Mohammad, Yasser},
title = {{[COMP25]} The Automated Negotiating Agents Competition {(ANAC)} 2025 Challenges and Results},
booktitle = {CoRR},
volume = {abs/2604.13914},
year = {2026},
doi = {10.48550/arXiv.2604.13914},
eprinttype = {arXiv},
eprint = {2604.13914},
url = {https://arxiv.org/abs/2604.13914}
}
Automated Negotiation
Mohammad, Y. (2025). Adapting Bargaining Strategies to the TAU Protocol for Better Negotiation Outcomes. ECAI 2025 — 28th European Conference on Artificial Intelligence, 413, 4977–4984.
@inproceedings{mohammad2025adapting,
keywords = {conference},
focus = {negotiation},
author = {Mohammad, Yasser},
title = {Adapting Bargaining Strategies to the {TAU} Protocol for Better Negotiation Outcomes},
booktitle = {ECAI 2025 --- 28th European Conference on Artificial Intelligence},
series = {Frontiers in Artificial Intelligence and Applications},
volume = {413},
pages = {4977--4984},
publisher = {IOS Press},
address = {Bologna, Italy},
year = {2025},
doi = {10.3233/FAIA251410}
}
As artificial intelligence (AI) becomes increasingly common in industry and business operations, researchers are focusing on how to enable AI agents representing different stakeholders to cooperate effectively in competitive environments. Inspired by human negotiation, automated negotiation is being explored as a solution. Traditional approaches like the Stacked Alternating Offers Protocol (SAOP) for multilateral negotiations and the Alternating Offers Protocol (AOP) for bilateral negotiations mimic human bargaining. Recently, the Tentative Acceptance Unique Offers Protocol (TAU) was proposed as an alternative that leads to faster and better agreements when a simple strategy is used by the negotiators. This paper introduces a method for adapting any existing AOP strategy to TAU. Our results demonstrate that this approach leads to faster and more beneficial agreements for all parties involved, with higher overall optimality and fairness. This improvement is achieved with a minor increase in information revelation.
Automated Negotiation
Ando, T., Miki, N., Yanagi, N., & Mohammad, Y. (2025). Automated Negotiation for Delivery Date Adjustment in Procurement. ECAI 2025 — 28th European Conference on Artificial Intelligence, 413, 5092–5095. Demo track.
@inproceedings{ando2025delivery,
keywords = {conference},
focus = {negotiation},
author = {Ando, Tomohito and Miki, Nozomoi and Yanagi, Norio and Mohammad, Yasser},
title = {Automated Negotiation for Delivery Date Adjustment in Procurement},
booktitle = {ECAI 2025 --- 28th European Conference on Artificial Intelligence},
series = {Frontiers in Artificial Intelligence and Applications},
volume = {413},
pages = {5092--5095},
publisher = {IOS Press},
address = {Bologna, Italy},
year = {2025},
doi = {10.3233/FAIA251427},
note = {Demo track}
}
Negotiation is ubiquitous in business applications in general and in procurement operations in particular. Nevertheless, several studies have shown that the negotiation process is often inefficient and time-consuming. In this paper, we propose a novel automated negotiation framework for procurement focusing on delivery date adjustment negotiations between buyers and suppliers. These negotiations are one of the most repeated negotiations in industrial applications.Nevertheless, they are often complex and time-consuming, as they involve multiple parties and require careful consideration of various internal and external factors. The proposed method was evaluated in the field and was shown to provide a significant reduction in the time required to reach achievement and around 95
Multiagent Systems
Mohammad, Y. (2025). Tackling the Protocol Problem in Automated Negotiation. Proceedings of the 24th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2025), 2870–2874. Blue Sky Ideas track. https://www.ifaamas.org/Proceedings/aamas2025/pdfs/p2870.pdf
@inproceedings{mohammad2025protocol,
keywords = {conference},
focus = {multiagent},
author = {Mohammad, Yasser},
title = {Tackling the Protocol Problem in Automated Negotiation},
booktitle = {Proceedings of the 24th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2025)},
pages = {2870--2874},
publisher = {IFAAMAS/ACM},
address = {Detroit, MI, USA},
year = {2025},
doi = {10.5555/3709347.3744037},
url = {https://www.ifaamas.org/Proceedings/aamas2025/pdfs/p2870.pdf},
note = {Blue Sky Ideas track}
}
Automated Negotiation (AN) is a research field with roots extending back to the mid-twentieth century. There are two dominant AN research directions pursued by the AAMAS community in recent years: (1) designing new heuristic or ML/RL/MARL-based strategies for the simplest bargaining mechanism called the Alternating Offers Protocol (AOP) and its extensions and (2) defining new mediated mechanisms that require a trusted third party. Intelligence lies in the strategy in the first direction and the mechanism in the latter. Either way, evaluation is almost always conducted in terms of empirical evaluation in some chosen set of negotiation scenarios. This paper argues for more efforts towards tackling the problem of in automated negotiation more rigorously by integrating ideas from mechanism-design literature. This requires, as a first step, a common language for expressing different negotiation protocols and strategies. We provide such a language which can represent a wide variety of negotiation protocols (both mediated and unmediated). We briefly outline our early effort in using this approach to provide a novel protocol with a provable Perfect Bayesian Equilibrium strategy that is also empirically effective.
Automated Negotiation
Aydoğan, R., Baarslag, T., Florijn, T. C. P., Fujita, K., Jonker, C. M., & Mohammad, Y. (2025). [COMP24] The Automated Negotiating Agents Competition (ANAC) 2024 Challenges and Results. Proceedings of the 24th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2025), 3000–3002. https://www.ifaamas.org/Proceedings/aamas2025/pdfs/p3000.pdf
@inproceedings{aydogan2025anac2024,
keywords = {conference},
focus = {negotiation},
author = {Aydo{\u{g}}an, Reyhan and Baarslag, Tim and Florijn, Tamara C. P. and Fujita, Katsuhide and Jonker, Catholijn M. and Mohammad, Yasser},
title = {{[COMP24]} The Automated Negotiating Agents Competition {(ANAC)} 2024 Challenges and Results},
booktitle = {Proceedings of the 24th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2025)},
pages = {3000--3002},
publisher = {IFAAMAS/ACM},
address = {Detroit, MI, USA},
year = {2025},
doi = {10.5555/3709347.3744072},
url = {https://www.ifaamas.org/Proceedings/aamas2025/pdfs/p3000.pdf}
}
Automated Negotiation
Mohammad, Y., Nakadai, S., & Greenwald, A. (2024). Automated Negotiation in Supply Chains: A Generalist Environment for RL/MARL Research. PRIMA 2024: Principles and Practice of Multi-Agent Systems — 25th International Conference, 15395, 19–24.
@inproceedings{mohammad2024generalist,
keywords = {conference},
focus = {negotiation},
author = {Mohammad, Yasser and Nakadai, Shinji and Greenwald, Amy},
title = {Automated Negotiation in Supply Chains: A Generalist Environment for {RL/MARL} Research},
booktitle = {PRIMA 2024: Principles and Practice of Multi-Agent Systems --- 25th International Conference},
series = {Lecture Notes in Computer Science},
volume = {15395},
pages = {19--24},
publisher = {Springer},
address = {Kyoto, Japan},
year = {2024},
doi = {10.1007/978-3-031-77367-9_2}
}
Automated Negotiation
Mohammad, Y. (2023). Generalized Bargaining Protocols. AI 2023: Advances in Artificial Intelligence — 36th Australasian Joint Conference on Artificial Intelligence, 14472, 261–273.
@inproceedings{mohammad2023generalizedbargaining,
keywords = {conference},
focus = {negotiation},
author = {Mohammad, Yasser},
title = {Generalized Bargaining Protocols},
booktitle = {AI 2023: Advances in Artificial Intelligence --- 36th Australasian Joint Conference on Artificial Intelligence},
series = {Lecture Notes in Computer Science},
volume = {14472},
pages = {261--273},
publisher = {Springer},
address = {Brisbane, Australia},
year = {2023},
doi = {10.1007/978-981-99-8391-9_21}
}
Automated Negotiation (AN) is a research field with roots extending back to the mid-twentieth century. There are two dominant AN research directions in recent years: (1) designing new heuristic strategies for the simplest bargaining protocol called the Alternating Offers Protocol (AOP) and (2) defining new mediated protocol that require a trusted third party. Intelligence lies in the strategy in the first direction and the protocol in the latter. This paper argues for a third way that aims at designing unmediated AN protocols with desired properties. We introduce a generalization of AOP to a wide class of unmediated protocols that keep its main advantages while providing the designer with the freedom to design protocols with desired properties. We also introduce the first fruits of this research direction in the form of an unmediated protocol and a corresponding simple strategy that can be shown theoretically to be exactly rational, optimal, and complete for bilateral negotiations with no information about partner’s preferences. Automated Negotiation Multiagent Systems Mechanism Design.
Automated Negotiation
Mohammad, Y. (2023). Evaluating Automated Negotiations. IEEE International Conference on Agents (ICA 2023), 77–82. Best Paper Award.
IEEE ICA 2023 Best Paper
@inproceedings{mohammad2023evaluating,
award = {IEEE ICA 2023 Best Paper},
keywords = {conference},
focus = {negotiation},
author = {Mohammad, Yasser},
title = {Evaluating Automated Negotiations},
booktitle = {IEEE International Conference on Agents (ICA 2023)},
pages = {77--82},
publisher = {IEEE},
address = {Kyoto, Japan},
year = {2023},
doi = {10.1109/ICA58824.2023.00022},
note = {Best Paper Award}
}
Automated Negotiation (AN) is a process for reaching agreement between agents representing self-interested parties. Several negotiation protocols and strategies have been proposed over the years. In this paper, we argue that evaluating a negotiation session is not straightforward and depends on several contextual factors. The paper proposes a multifaceted evaluation criteria for negotiation sessions that can be adjusted to the evaluation context. The proposed evaluation criteria are then applied to several negotiation scenarios with state-of-the-art negotiation algorithms.
Automated Negotiation
Aydoğan, R., Baarslag, T., Fujita, K., Hoos, H. H., Jonker, C. M., Mohammad, Y., & Renting, B. M. (2022). The 13th International Automated Negotiating Agent Competition Challenges and Results. Recent Advances in Agent-Based Negotiation: Applications and Competition Challenges (ACAN@IJCAI 2022), 1092, 87–101.
@inproceedings{aydogan2022anac13,
keywords = {conference},
focus = {negotiation},
author = {Aydo{\u{g}}an, Reyhan and Baarslag, Tim and Fujita, Katsuhide and Hoos, Holger H. and Jonker, Catholijn M. and Mohammad, Yasser and Renting, Bram M.},
title = {The 13th International Automated Negotiating Agent Competition Challenges and Results},
booktitle = {Recent Advances in Agent-Based Negotiation: Applications and Competition Challenges (ACAN@IJCAI 2022)},
series = {Studies in Computational Intelligence},
volume = {1092},
pages = {87--101},
publisher = {Springer},
year = {2022},
doi = {10.1007/978-981-99-0561-4_5}
}
Automated Negotiation
Mohammad, Y., & Nakadai, S. (2022). Concurrent Negotiations with Global Utility Functions. Proceedings of the 21st International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2022), 1947–1949. https://www.ifaamas.org/Proceedings/aamas2022/pdfs/p1947.pdf
@inproceedings{mohammad2022concurrentglobal,
keywords = {conference, important},
focus = {negotiation},
author = {Mohammad, Yasser and Nakadai, Shinji},
title = {Concurrent Negotiations with Global Utility Functions},
booktitle = {Proceedings of the 21st International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2022)},
pages = {1947--1949},
publisher = {IFAAMAS},
address = {Auckland, New Zealand},
year = {2022},
doi = {10.5555/3535850.3536162},
url = {https://www.ifaamas.org/Proceedings/aamas2022/pdfs/p1947.pdf}
}
Automated Negotiation is attracting more attention from researchers recently as it is becoming more relevant to industrial and business applications with increased reliance on automated systems. Most research in this area assumes either a single negotiation thread with a well-defined utility function for each agent involved or a set of concurrent negotiations with an ordering of outcomes in each local negotiation. In this paper, we consider an agent engaged in a set of concurrent negotiations with a utility function defined only for the in of them and no locally defined ordering of outcomes in any negotiation independent from what happens in the others. We argue that this problem setting is interesting both from the academic and the industrial points of view. The paper then presents an algorithm that allows such agent to maximize its expected global utility by orchestrating its behavior in all negotiation threads. The performance of the proposed method is analyzed theoretically and empirically using simulation.
Automated Negotiation
Sengupta, A., Nakadai, S., & Mohammad, Y. (2022). Transfer Learning Based Adaptive Automated Negotiating Agent Framework. Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence (IJCAI 2022), 468–474. https://www.ijcai.org/proceedings/2022/0067.pdf
@inproceedings{sengupta2022transfer,
keywords = {conference},
focus = {negotiation},
author = {Sengupta, Ayan and Nakadai, Shinji and Mohammad, Yasser},
title = {Transfer Learning Based Adaptive Automated Negotiating Agent Framework},
booktitle = {Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence (IJCAI 2022)},
pages = {468--474},
publisher = {ijcai.org},
address = {Vienna, Austria},
year = {2022},
doi = {10.24963/ijcai.2022/67},
url = {https://www.ijcai.org/proceedings/2022/0067.pdf}
}
ML & Time-Series
Heracleous, P., Fukayama, S., Ogata, J., & Mohammad, Y. (2022). Applying Generative Adversarial Networks and Vision Transformers in Speech Emotion Recognition. HCI International 2022 — Late Breaking Papers (HCII 2022), 13519, 67–75.
@inproceedings{heracleous2022gan,
keywords = {conference},
focus = {timeseries},
author = {Heracleous, Panikos and Fukayama, Satoru and Ogata, Jun and Mohammad, Yasser},
title = {Applying Generative Adversarial Networks and Vision Transformers in Speech Emotion Recognition},
booktitle = {HCI International 2022 --- Late Breaking Papers (HCII 2022)},
series = {Lecture Notes in Computer Science},
volume = {13519},
pages = {67--75},
publisher = {Springer},
year = {2022},
doi = {10.1007/978-3-031-17618-0_6}
}
Robotics & HRI
Ahmed, A., Mohammad, Y. F. O., Parque, V., El-Hussieny, H., & Ahmed, S. M. (2022). End-to-End Mobile Robot Navigation Using a Residual Deep Reinforcement Learning in Dynamic Human Environments. 18th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications (MESA 2022), 1–6.
@inproceedings{ahmed2022endtoend,
keywords = {conference},
focus = {robotics},
author = {Ahmed, Abdullah and Mohammad, Yasser F. O. and Parque, Victor and El-Hussieny, Haitham and Ahmed, Sabah M.},
title = {End-to-End Mobile Robot Navigation Using a Residual Deep Reinforcement Learning in Dynamic Human Environments},
booktitle = {18th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications (MESA 2022)},
pages = {1--6},
publisher = {IEEE},
address = {Taipei, Taiwan},
year = {2022},
doi = {10.1109/MESA55290.2022.10004394}
}
Automated Negotiation
Sengupta, A., Mohammad, Y., & Nakadai, S. (2021). An Autonomous Negotiating Agent Framework with Reinforcement Learning Based Strategies and Adaptive Strategy Switching Mechanism. Proceedings of the 20th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2021), 1163–1172. https://ifaamas.org/Proceedings/aamas2021/pdfs/p1163.pdf
@inproceedings{sengupta2021autonomous,
keywords = {conference, important},
focus = {negotiation},
author = {Sengupta, Ayan and Mohammad, Yasser and Nakadai, Shinji},
title = {An Autonomous Negotiating Agent Framework with Reinforcement Learning Based Strategies and Adaptive Strategy Switching Mechanism},
booktitle = {Proceedings of the 20th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2021)},
pages = {1163--1172},
publisher = {ACM},
year = {2021},
doi = {10.5555/3463952.3464087},
url = {https://ifaamas.org/Proceedings/aamas2021/pdfs/p1163.pdf}
}
Despite abundant negotiation strategies in literature, the complexity of automated negotiation forbids a single strategy from being dominant against all others in different negotiation scenarios. To overcome this, one approach is to use mixture of experts, but at the same time one problem of this method is the selection of experts, as this approach is limited by the competency of the experts selected. Another problem with most negotiation strategies is their incapability of adapting to dynamic variation of the opponent’s behaviour within a single negotiation session resulting in poor performance. This work focuses on both, solving the problem of expert selection and adapting to the opponent’s behaviour with our Autonomous Negotiating Agent Framework. This framework allows real-time classification of opponent’s behaviour and provides a mechanism to select, switch or combine strategies within a single negotiation session. Additionally, our framework has a reviewer component which enables self-enhancement capability by deciding to include new strategies or replace old ones with better strategies periodically. We demonstrate an instance of our framework by implementing maximum entropy reinforcement learning based strategies with a deep learning based opponent classifier. Finally, we evaluate the performance of our agent against state-of-the-art negotiators under varied negotiation scenarios.
ML & Time-Series
Heracleous, P., Mohammad, Y., & Yoneyama, A. (2021). Speech Emotion Recognition Using Combined Multiple Pairwise Classifiers. HCI International 2021 — Late Breaking Posters (HCII 2021), 1498, 115–118.
@inproceedings{heracleous2021pairwise,
keywords = {conference},
focus = {timeseries},
author = {Heracleous, Panikos and Mohammad, Yasser and Yoneyama, Akio},
title = {Speech Emotion Recognition Using Combined Multiple Pairwise Classifiers},
booktitle = {HCI International 2021 --- Late Breaking Posters (HCII 2021)},
series = {Communications in Computer and Information Science},
volume = {1498},
pages = {115--118},
publisher = {Springer},
year = {2021},
doi = {10.1007/978-3-030-90176-9_16}
}
Automated Negotiation
Mohammad, Y., Nakadai, S., & Greenwald, A. (2020). NegMAS: A Platform for Automated Negotiations. PRIMA 2020: Principles and Practice of Multi-Agent Systems — 23rd International Conference, 12568, 343–351.
@inproceedings{mohammad2020negmas,
keywords = {conference},
focus = {negotiation},
author = {Mohammad, Yasser and Nakadai, Shinji and Greenwald, Amy},
title = {{NegMAS}: A Platform for Automated Negotiations},
booktitle = {PRIMA 2020: Principles and Practice of Multi-Agent Systems --- 23rd International Conference},
series = {Lecture Notes in Computer Science},
volume = {12568},
pages = {343--351},
publisher = {Springer},
address = {Nagoya, Japan},
year = {2020},
doi = {10.1007/978-3-030-69322-0_23}
}
Alongside the widespread adoption of AI technology thoughout the business world, automated negotiation is similarly gaining more interest within the multiagent system (MAS) research community. This interest has prompted the development of research-oriented automated negotiation platforms like GENIUS. This paper introduces NegMAS, Negotiations Managed by Agent Simulations / Negotiation MultiAgent System, which was developed to facilitate research and development of agents that negotiate in dynamic situations characterized by interrelated utility functions with all negotiation related decisions managed by agents.
Automated Negotiation
Aydoğan, R., Baarslag, T., Fujita, K., Mell, J., Gratch, J., de Jonge, D., Mohammad, Y., Nakadai, S., Morinaga, S., Osawa, H., Aranha, C., & Jonker, C. M. (2020). Challenges and Main Results of the Automated Negotiating Agents Competition (ANAC) 2019. Multi-Agent Systems and Agreement Technologies — 17th European Conference (EUMAS 2020) and 7th International Conference (AT 2020), 12520, 366–381.
@inproceedings{aydogan2020anac2019,
keywords = {conference},
focus = {negotiation},
author = {Aydo{\u{g}}an, Reyhan and Baarslag, Tim and Fujita, Katsuhide and Mell, Johnathan and Gratch, Jonathan and de Jonge, Dave and Mohammad, Yasser and Nakadai, Shinji and Morinaga, Satoshi and Osawa, Hirotaka and Aranha, Claus and Jonker, Catholijn M.},
title = {Challenges and Main Results of the Automated Negotiating Agents Competition {(ANAC)} 2019},
booktitle = {Multi-Agent Systems and Agreement Technologies --- 17th European Conference (EUMAS 2020) and 7th International Conference (AT 2020)},
series = {Lecture Notes in Computer Science},
volume = {12520},
pages = {366--381},
publisher = {Springer},
address = {Thessaloniki, Greece},
year = {2020},
doi = {10.1007/978-3-030-66412-1_23}
}
ML & Time-Series
Heracleous, P., Mohammad, Y., & Yoneyama, A. (2020). Integrating Language and Emotion Features for Multilingual Speech Emotion Recognition. Human-Computer Interaction. Multimodal and Natural Interaction (HCII 2020), 12182, 187–196.
@inproceedings{heracleous2020multilingual,
keywords = {conference},
focus = {timeseries},
author = {Heracleous, Panikos and Mohammad, Yasser and Yoneyama, Akio},
title = {Integrating Language and Emotion Features for Multilingual Speech Emotion Recognition},
booktitle = {Human-Computer Interaction. Multimodal and Natural Interaction (HCII 2020)},
series = {Lecture Notes in Computer Science},
volume = {12182},
pages = {187--196},
publisher = {Springer},
address = {Copenhagen, Denmark},
year = {2020},
doi = {10.1007/978-3-030-49062-1_12}
}
ML & Time-Series
Heracleous, P., Takai, K., Wang, Y., Yasuda, K., Yoneyama, A., & Mohammad, Y. (2020). An Empirical Study on Feature Extraction in DNN-Based Speech Emotion Recognition. HCI International 2020 — Late Breaking Posters (HCII 2020), 1293, 315–319.
@inproceedings{heracleous2020empirical,
keywords = {conference},
focus = {timeseries},
author = {Heracleous, Panikos and Takai, Kohichi and Wang, Yanan and Yasuda, Keiji and Yoneyama, Akio and Mohammad, Yasser},
title = {An Empirical Study on Feature Extraction in {DNN}-Based Speech Emotion Recognition},
booktitle = {HCI International 2020 --- Late Breaking Posters (HCII 2020)},
series = {Communications in Computer and Information Science},
volume = {1293},
pages = {315--319},
publisher = {Springer},
address = {Copenhagen, Denmark},
year = {2020},
doi = {10.1007/978-3-030-60700-5_40}
}
Multiagent Systems
Mohammad, Y. (2020). Optimal Deterministic Time-Based Policy in Automated Negotiation. PRIMA 2020: Principles and Practice of Multi-Agent Systems — 23rd International Conference, 12568, 68–83. Best Paper Award runner-up.
PRIMA 2020 Best Paper Runner-up
@inproceedings{mohammad2020optimaldeterministic,
award = {PRIMA 2020 Best Paper Runner-up},
keywords = {conference},
focus = {multiagent},
author = {Mohammad, Yasser},
title = {Optimal Deterministic Time-Based Policy in Automated Negotiation},
booktitle = {PRIMA 2020: Principles and Practice of Multi-Agent Systems --- 23rd International Conference},
series = {Lecture Notes in Computer Science},
volume = {12568},
pages = {68--83},
publisher = {Springer},
address = {Nagoya, Japan},
year = {2020},
doi = {10.1007/978-3-030-69322-0_5},
note = {Best Paper Award runner-up}
}
Automated negotiation is gaining more attention as a possible mechanism for organizing self-interested intelligent agents in a distributed environment. The problem of designing effective negotiation strategies in such environments was studied extensively by researchers from economics, computer science, multiagent systems, and AI. This paper focuses on the problem of finding effective deterministic time-based offering strategies given an opponent acceptance model. This problem was studied earlier and optimal solutions are known for the simplest case of a static stationary acceptance model. This paper proposes an efficient approach for calculating the effect of different manipulations of the offering policy on expected utility and uses that to provide a faster implementation of the optimal algorithm for static stationary acceptance models and provide an approximate extension to more realistic acceptance models.
ML & Time-Series
Heracleous, P., Mohammad, Y., Yasuda, K., & Yoneyama, A. (2019). Speech Emotion Recognition Using Spontaneous Children’s Corpus. Computational Linguistics and Intelligent Text Processing — 20th International Conference (CICLing 2019), 13452, 321–333.
@inproceedings{heracleous2019children,
keywords = {conference},
focus = {timeseries},
author = {Heracleous, Panikos and Mohammad, Yasser and Yasuda, Keiji and Yoneyama, Akio},
title = {Speech Emotion Recognition Using Spontaneous Children's Corpus},
booktitle = {Computational Linguistics and Intelligent Text Processing --- 20th International Conference (CICLing 2019)},
series = {Lecture Notes in Computer Science},
volume = {13452},
pages = {321--333},
publisher = {Springer},
address = {La Rochelle, France},
year = {2019},
doi = {10.1007/978-3-031-24340-0_24}
}
Automatic recognition of human emotions is a relatively new field and is attracting significant attention in research and development areas because of the major contribution it could make to real applications. Previously, several studies reported speech emotion recognition using acted emotional corpus. For real world applications, however, spontaneous corpora should be used in recognizing human emotions from speech. This study focuses on speech emotion recognition using the FAU Aibo spontaneous children’s corpus. A method based on the integration of feed-forward deep neural networks (DNN) and the i-vector paradigm is proposed, and another method based on deep convolutional neural networks (DCNN) for feature extraction and extremely randomized trees as classifier is presented. For the classification of five emotions using balanced data, the proposed methods showed unweighted average recalls (UAR) of 61.1
ML & Time-Series
Heracleous, P., Mohammad, Y., & Yoneyama, A. (2019). Deep Convolutional Neural Networks for Feature Extraction in Speech Emotion Recognition. Human-Computer Interaction. Recognition and Interaction Technologies (HCII 2019), 11567, 117–132.
@inproceedings{heracleous2019deepcnn,
keywords = {conference},
focus = {timeseries},
author = {Heracleous, Panikos and Mohammad, Yasser and Yoneyama, Akio},
title = {Deep Convolutional Neural Networks for Feature Extraction in Speech Emotion Recognition},
booktitle = {Human-Computer Interaction. Recognition and Interaction Technologies (HCII 2019)},
series = {Lecture Notes in Computer Science},
volume = {11567},
pages = {117--132},
publisher = {Springer},
address = {Orlando, FL, USA},
year = {2019},
doi = {10.1007/978-3-030-22643-5_9}
}
ML & Time-Series
Rayan, Y., Mohammad, Y. F. O., & Ali, S. A. (2019). Multidimensional Permutation Entropy for Constrained Motif Discovery. Intelligent Information and Database Systems — 11th Asian Conference (ACIIDS 2019), 11431, 231–243.
@inproceedings{rayan2019multidimensional,
keywords = {conference},
focus = {timeseries},
author = {Rayan, Yomna and Mohammad, Yasser F. O. and Ali, Samia A.},
title = {Multidimensional Permutation Entropy for Constrained Motif Discovery},
booktitle = {Intelligent Information and Database Systems --- 11th Asian Conference (ACIIDS 2019)},
series = {Lecture Notes in Computer Science},
volume = {11431},
pages = {231--243},
publisher = {Springer},
address = {Yogyakarta, Indonesia},
year = {2019},
doi = {10.1007/978-3-030-14799-0_20}
}
Automated Negotiation
Mohammad, Y., Areyan Viqueira, E., Ayerza, N. A., Greenwald, A., Nakadai, S., & Morinaga, S. (2019). Supply Chain Management World — A Benchmark Environment for Situated Negotiations. PRIMA 2019: Principles and Practice of Multi-Agent Systems — 22nd International Conference, 11873, 153–169.
@inproceedings{mohammad2019scmworld,
keywords = {conference},
focus = {negotiation},
author = {Mohammad, Yasser and {Areyan Viqueira}, Enrique and Ayerza, Nahum Alvarez and Greenwald, Amy and Nakadai, Shinji and Morinaga, Satoshi},
title = {Supply Chain Management World --- A Benchmark Environment for Situated Negotiations},
booktitle = {PRIMA 2019: Principles and Practice of Multi-Agent Systems --- 22nd International Conference},
series = {Lecture Notes in Computer Science},
volume = {11873},
pages = {153--169},
publisher = {Springer},
address = {Turin, Italy},
year = {2019},
doi = {10.1007/978-3-030-33792-6_10}
}
Automated Negotiation
Areyan Viqueira, E., Cousins, C., Mohammad, Y., & Greenwald, A. (2019). Empirical Mechanism Design: Designing Mechanisms from Data. Proceedings of the Thirty-Fifth Conference on Uncertainty in Artificial Intelligence (UAI 2019), 115, 1094–1104. https://proceedings.mlr.press/v115/viqueira20a.html
@inproceedings{viqueira2019empirical,
keywords = {conference},
focus = {negotiation},
author = {{Areyan Viqueira}, Enrique and Cousins, Cyrus and Mohammad, Yasser and Greenwald, Amy},
title = {Empirical Mechanism Design: Designing Mechanisms from Data},
booktitle = {Proceedings of the Thirty-Fifth Conference on Uncertainty in Artificial Intelligence (UAI 2019)},
series = {Proceedings of Machine Learning Research},
volume = {115},
pages = {1094--1104},
publisher = {AUAI Press},
address = {Tel Aviv, Israel},
year = {2019},
url = {https://proceedings.mlr.press/v115/viqueira20a.html}
}
Automated Negotiation
Mohammad, Y. F. O., & Nakadai, S. (2019). Optimal Value of Information Based Elicitation During Negotiation. Proceedings of the 18th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2019), 242–250. https://www.ifaamas.org/Proceedings/aamas2019/pdfs/p242.pdf
@inproceedings{mohammad2019voi,
keywords = {conference},
focus = {negotiation},
author = {Mohammad, Yasser F. O. and Nakadai, Shinji},
title = {Optimal Value of Information Based Elicitation During Negotiation},
booktitle = {Proceedings of the 18th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2019)},
pages = {242--250},
publisher = {IFAAMAS},
address = {Montreal, QC, Canada},
year = {2019},
url = {https://www.ifaamas.org/Proceedings/aamas2019/pdfs/p242.pdf}
}
Autonomous agents engaging in automatic negotiations on behalf of humans or institutions are usually assumed to have full knowledge of the utility function for the actors they represent. In many cases, these utility functions are difficult to know apriori for every possible outcome of the negotiation. Moreover, it may not be necessary for the agent to know the utility of outcomes that are never offered or considered during the negotiation. State-of-the-art approaches to utility elicitation during negotiation assume that the agent can ask from a predefined countable set to reduce its uncertainty about the utility function. This paper extends that body of work by lifting the countability assumption providing an optimal algorithm for selecting the best outcome and utility level about which to ask the actor. The paper reports the results of comparing the proposed algorithm with state-of-the-art algorithms using both synthetic and realistic negotiation scenarios. These evaluations support the applicability of the proposed approach.
Automated Negotiation
Mohammad, Y. F. O., & Nakadai, S. (2018). Utility Elicitation During Negotiation with Practical Elicitation Strategies. IEEE International Conference on Systems, Man, and Cybernetics (SMC 2018), 3100–3107.
@inproceedings{mohammad2018elicitation,
keywords = {conference},
focus = {negotiation},
author = {Mohammad, Yasser F. O. and Nakadai, Shinji},
title = {Utility Elicitation During Negotiation with Practical Elicitation Strategies},
booktitle = {IEEE International Conference on Systems, Man, and Cybernetics (SMC 2018)},
pages = {3100--3107},
publisher = {IEEE},
address = {Miyazaki, Japan},
year = {2018},
doi = {10.1109/SMC.2018.00525}
}
Automatic negotiation is gaining more interest recently thanks to the wider deployment of intelligent systems and the need for them to cooperate/compete on behalf of their users. A central assumption of most autonomous negotiation agents is that the utility function of the user is perfectly known to the agent. That is an often unmet assumption in real situations. Utility elicitation is the process of learning about the utility function of the user incrementally and has a long history in decision support research. Recently, some utility elicitation systems capable of incrementally eliciting the utility function of the user during the negotiation were presented. This work expands this body of research by optimizing the elicitation algorithm to realistic elicitation strategies. The proposed method extends the optimal elicitation algorithm to the – practical – case where queries to the user only reduce the uncertainty in the utility function without removing it completely. Extensive evaluation shows that the proposed extension outperforms two state-of-the-art elicitation algorithms and several baseline alternatives.
ML & Time-Series
Heracleous, P., Mohammad, Y., Takai, K., Yasuda, K., & Yoneyama, A. (2018). Spoken Language Identification Based on I-Vectors and Conditional Random Fields. 14th International Wireless Communications & Mobile Computing Conference (IWCMC 2018), 1443–1447.
@inproceedings{heracleous2018crf,
keywords = {conference},
focus = {timeseries},
author = {Heracleous, Panikos and Mohammad, Yasser and Takai, Kohichi and Yasuda, Keiji and Yoneyama, Akio},
title = {Spoken Language Identification Based on I-Vectors and Conditional Random Fields},
booktitle = {14th International Wireless Communications \& Mobile Computing Conference (IWCMC 2018)},
pages = {1443--1447},
publisher = {IEEE},
address = {Limassol, Cyprus},
year = {2018},
doi = {10.1109/IWCMC.2018.8450327}
}
ML & Time-Series
Heracleous, P., Mohammad, Y., Takai, K., Yasuda, K., & Yoneyama, A. (2018). I-Vectors and Deep Convolutional Neural Networks for Language Identification in Clean and Reverberant Environments. Computational Linguistics and Intelligent Text Processing — 19th International Conference (CICLing 2018), 13396, 30–40.
@inproceedings{heracleous2018ivectors,
keywords = {conference},
focus = {timeseries},
author = {Heracleous, Panikos and Mohammad, Yasser and Takai, Kohichi and Yasuda, Keiji and Yoneyama, Akio},
title = {I-Vectors and Deep Convolutional Neural Networks for Language Identification in Clean and Reverberant Environments},
booktitle = {Computational Linguistics and Intelligent Text Processing --- 19th International Conference (CICLing 2018)},
series = {Lecture Notes in Computer Science},
volume = {13396},
pages = {30--40},
publisher = {Springer},
address = {Hanoi, Vietnam},
year = {2018},
doi = {10.1007/978-3-031-23793-5_3}
}
In the current study, a method for automatic language identification based on deep convolutional neural networks (DCNN) and the i-vector paradigm is proposed. Convolutional neural networks (CNN) have been successfully applied to image classification, speech emotion recognition, and facial expression recognition. In the current study, a variant of typical CNN is being applied and experimentally investigated in spoken language identification. When the proposed method was evaluated on the NIST 2015 i-vector Machine Learning Challenge task for the recognition of 50 in-set languages, a 3.9
ML & Time-Series
Heracleous, P., Mohammad, Y., Takai, K., Yasuda, K., Yoneyama, A., & Sugaya, F. (2018). A Study on Far-Field Emotion Recognition Based on Deep Convolutional Neural Networks. Computational Linguistics and Intelligent Text Processing — 19th International Conference (CICLing 2018), 13397, 181–193.
@inproceedings{heracleous2018farfield,
keywords = {conference},
focus = {timeseries},
author = {Heracleous, Panikos and Mohammad, Yasser and Takai, Kohichi and Yasuda, Keiji and Yoneyama, Akio and Sugaya, Fumiaki},
title = {A Study on Far-Field Emotion Recognition Based on Deep Convolutional Neural Networks},
booktitle = {Computational Linguistics and Intelligent Text Processing --- 19th International Conference (CICLing 2018)},
series = {Lecture Notes in Computer Science},
volume = {13397},
pages = {181--193},
publisher = {Springer},
address = {Hanoi, Vietnam},
year = {2018},
doi = {10.1007/978-3-031-23804-8_15}
}
Automatic recognition of human emotions is a relatively new field, and is attracting significant attention in research and development areas because of the major contribution it could make to real applications. The current study focuses on far-field speech emotion recognition using the state-of-the-art spontaneous IEMOCAP emotional data. For classification, a method based on deep convolutional neural networks (DCNN) and extremely randomized trees is proposed. The method is also compared to support vector machines (SVM) and probabilistic linear discriminant analysis (PLDA) classifiers in the i-vector paradigm. When reverberant speech was classified using the proposed method, the classification rates were comparable to those obtained when using clean data. In the case of PLDA and SVM classifiers, the classification rates were significantly decreased. To further improve the performance of far-field speech emotion recognition, a method based on multi-style training is proposed, which results in significant improvements in the classification rates.
ML & Time-Series
Heracleous, P., Takai, K., Yasuda, K., Mohammad, Y., & Yoneyama, A. (2018). Comparative Study on Spoken Language Identification Based on Deep Learning. 26th European Signal Processing Conference (EUSIPCO 2018), 2265–2269.
@inproceedings{heracleous2018comparative,
keywords = {conference},
focus = {timeseries},
author = {Heracleous, Panikos and Takai, Kohichi and Yasuda, Keiji and Mohammad, Yasser and Yoneyama, Akio},
title = {Comparative Study on Spoken Language Identification Based on Deep Learning},
booktitle = {26th European Signal Processing Conference (EUSIPCO 2018)},
pages = {2265--2269},
publisher = {IEEE},
address = {Rome, Italy},
year = {2018},
doi = {10.23919/EUSIPCO.2018.8553347}
}
Automated Negotiation
Mohammad, Y. F. O., & Nakadai, S. (2018). FastVOI: Efficient Utility Elicitation During Negotiations. PRIMA 2018: Principles and Practice of Multi-Agent Systems — 21st International Conference, 11224, 560–567.
@inproceedings{mohammad2018fastvoi,
keywords = {conference},
focus = {negotiation},
author = {Mohammad, Yasser F. O. and Nakadai, Shinji},
title = {{FastVOI}: Efficient Utility Elicitation During Negotiations},
booktitle = {PRIMA 2018: Principles and Practice of Multi-Agent Systems --- 21st International Conference},
series = {Lecture Notes in Computer Science},
volume = {11224},
pages = {560--567},
publisher = {Springer},
address = {Tokyo, Japan},
year = {2018},
doi = {10.1007/978-3-030-03098-8_42}
}
Autonomous Negotiation is a promising technology that allows individuals and institutions to reduce the burden and cost of negotiating win-win agreements. A common challenge in practical applications is the inability or high cost of finding the utility value for each possible outcome of the negotiation before it even starts. Earlier work on utility elicitation during negotiations tried to avoid the need of full revelation of the utility function to the agent by interleaving elicitation and negotiation actions. This paper proposes an efficient elicitation algorithm that allows the agent to achieve similar utility at orders of magnitude higher speed compared with the state-of-the-art algorithm. Autonomous Negotiation Utility Elicitation.
ML & Time-Series
Mohammad, Y. F. O., Matsumoto, K., & Hoashi, K. (2018). Deep Feature Learning and Selection for Activity Recognition. Proceedings of the 33rd Annual ACM Symposium on Applied Computing (SAC 2018), 930–939.
@inproceedings{mohammad2018deepfeature,
keywords = {conference},
focus = {timeseries},
author = {Mohammad, Yasser F. O. and Matsumoto, Kazunori and Hoashi, Keiichiro},
title = {Deep Feature Learning and Selection for Activity Recognition},
booktitle = {Proceedings of the 33rd Annual ACM Symposium on Applied Computing (SAC 2018)},
pages = {930--939},
publisher = {ACM},
address = {Pau, France},
year = {2018},
doi = {10.1145/3167132.3167234}
}
Human physical activity recognition from sensor data is a growing area of research due to the widespread adoption of sensor-rich wearable and smart devices. The growing interest resulted in several formulations with multiple proposals for each of them. This paper is interested in activity recognition from short sequences of sensor readings. Traditionally, solutions to this problem have relied on handcrafted features and feature selection from large predefined feature sets. More recently, deep methods have been employed to provide an end-to-end classification system for activity recognition with higher accuracy at the expense of much slower performance. This paper proposes a middle ground in which a deep neural architecture is employed for feature learning followed by traditional feature selection and classification. This approach is shown to outperform state-of-the-art systems on six out of seven experiments using publicly available datasets.
ML & Time-Series
Mohammad, Y. F. O., Matsumoto, K., & Hoashi, K. (2017). A Dataset for Activity Recognition in an Unmodified Kitchen Using Smart-Watch Accelerometers. Proceedings of the 16th International Conference on Mobile and Ubiquitous Multimedia (MUM 2017), 63–68.
@inproceedings{mohammad2017kitchen,
keywords = {conference},
focus = {timeseries},
author = {Mohammad, Yasser F. O. and Matsumoto, Kazunori and Hoashi, Keiichiro},
title = {A Dataset for Activity Recognition in an Unmodified Kitchen Using Smart-Watch Accelerometers},
booktitle = {Proceedings of the 16th International Conference on Mobile and Ubiquitous Multimedia (MUM 2017)},
pages = {63--68},
publisher = {ACM},
address = {Stuttgart, Germany},
year = {2017},
doi = {10.1145/3152832.3152844}
}
Robotics & HRI
Hussein, M., Mohammad, Y., Ali, S. A., & Nishida, T. (2017). COLD: A ROS Package for Continuous Learning from Demonstration: Teaching a Robot to Write. IEEE International Conference on Mechatronics and Automation (ICMA 2017), 651–657.
@inproceedings{hussein2017cold,
keywords = {conference},
focus = {robotics},
author = {Hussein, Mostafa and Mohammad, Yasser and Ali, Samia A. and Nishida, Toyoaki},
title = {{COLD}: A {ROS} Package for Continuous Learning from Demonstration: Teaching a Robot to Write},
booktitle = {IEEE International Conference on Mechatronics and Automation (ICMA 2017)},
pages = {651--657},
publisher = {IEEE},
address = {Takamatsu, Japan},
year = {2017}
}
Learning From Demonstration (LfD) is an important area of research in robotics because it provides the means for skill transfer from humans to robots with no reliance on any specific technical skills of the human teacher. There are many approaches to LfD. Most available methods assume the existence of clear boundaries between different demonstrations. Less research focuses on techniques that can be used by the robot to learn from continuous streams of data that provide not only planned but also unplanned demonstrations. This paper reports an integrated framework to tackle this problem using the Robot Operating System (ROS) that is modular, scalable, easily adaptable, and robot-independent. We also report a study applying this method to teach robots how to write in both English and Arabic by just watching people writing short reports. Not only can the learner write words that exist in the training data but novel words based on the basic strokes learned from the demonstrations. The system was applied to teaching both a physical NAO robot and a Parallel robot simulator and evaluations demonstrate the effectiveness of the proposed framework in this task.
ML & Time-Series
Mohammad, Y. F. O., & Nishida, T. (2016). MC\^2: An Integrated Toolbox for Change, Causality and Motif Discovery. Trends in Applied Knowledge-Based Systems and Data Science — 29th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2016), 9799, 128–141.
@inproceedings{mohammad2016mc2,
keywords = {conference},
focus = {timeseries},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {{MC\^{}2}: An Integrated Toolbox for Change, Causality and Motif Discovery},
booktitle = {Trends in Applied Knowledge-Based Systems and Data Science --- 29th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2016)},
series = {Lecture Notes in Computer Science},
volume = {9799},
pages = {128--141},
publisher = {Springer},
address = {Morioka, Japan},
year = {2016},
doi = {10.1007/978-3-319-42007-3_12}
}
Time series are being generated continuously from all kinds of human endeavors. The ubiquity of time-series data generates a need for data mining and pattern discovery algorithms targeting this data format which is becoming of ever increasing importance. Three basic problems in mining time-series data are change point discovery, causality discovery and motif discovery. This paper presents an integrated toolbox that can be used to perform any of these tasks on multidimensional real-valued time-series using state of the art algorithms. The proposed toolbox provides practitioners in time-series analysis and data mining with several tools useful for data generation, preprocessing, modeling evaluation and mining of long sequences. As a use–case of the toolbox, we provide a comparison between three variants of the GEMODA algorithm applied to real–valued time-series data and a state-of-the-art stochastic MD algorithm. The paper also reports real world applications that uses the toolbox in HRI, physiological signal processing, and human behavior modeling and understanding. Change Point Discovery, Motif Discovery, Causality Analysis
Robotics & HRI
Hussein, M., Mohammad, Y., & Ali, S. A. (2015). Learning from Demonstration Using Variational Bayesian Inference. 28th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2015), 371–381.
@inproceedings{hussein2015variational,
keywords = {conference},
focus = {robotics},
author = {Hussein, Mostafa and Mohammad, Yasser and Ali, Samia A.},
title = {Learning from Demonstration Using Variational Bayesian Inference},
booktitle = {28th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2015)},
pages = {371--381},
publisher = {Springer},
address = {Seoul, South Korea},
year = {2015}
}
Nishida, T., Abe, M., Ookaki, T., Lala, D., Thovutikul, S., Song, H., Mohammad, Y. F. O., Nitschke, C., Ohmoto, Y., Nakazawa, A., Shochi, T., Rouas, J.-L., Bugeau, A., Lotte, F., Ming, Z., Letournel, G., Guerry, M., & Fourer, D. (2015). Synthetic Evidential Study as Augmented Collective Thought Process — Preliminary Report. Intelligent Information and Database Systems — 7th Asian Conference (ACIIDS 2015), 9011, 13–22.
@inproceedings{nishida2015synthetic,
keywords = {conference},
author = {Nishida, Toyoaki and Abe, Masakazu and Ookaki, Takashi and Lala, Divesh and Thovutikul, Sutasinee and Song, Hengjie and Mohammad, Yasser F. O. and Nitschke, Christian and Ohmoto, Yoshimasa and Nakazawa, Atsushi and Shochi, Takaaki and Rouas, Jean-Luc and Bugeau, Aur{\'e}lie and Lotte, Fabien and Ming, Zuheng and Letournel, Geoffrey and Guerry, Marine and Fourer, Dominique},
title = {Synthetic Evidential Study as Augmented Collective Thought Process --- Preliminary Report},
booktitle = {Intelligent Information and Database Systems --- 7th Asian Conference (ACIIDS 2015)},
series = {Lecture Notes in Computer Science},
volume = {9011},
pages = {13--22},
publisher = {Springer},
address = {Bali, Indonesia},
year = {2015},
doi = {10.1007/978-3-319-15702-3_2}
}
Robotics & HRI
Mohammad, Y., & Nishida, T. (2015). Simple Incremental GMM Modeling Using Multidimensional Piecewise Linear Segmentation for Learning from Demonstration. 3rd International Conference on Industrial Application Engineering (ICIAE 2015). Best Presentation Award.
ICIAE 2015 Best Presentation
@inproceedings{mohammad2015gmm,
award = {ICIAE 2015 Best Presentation},
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser and Nishida, Toyoaki},
title = {Simple Incremental {GMM} Modeling Using Multidimensional Piecewise Linear Segmentation for Learning from Demonstration},
booktitle = {3rd International Conference on Industrial Application Engineering (ICIAE 2015)},
address = {Kitakyushu, Japan},
year = {2015},
note = {Best Presentation Award}
}
ML & Time-Series
Mohammad, Y. F. O., & Nishida, T. (2014). Exact Discovery of Length-Range Motifs. Intelligent Information and Database Systems — 6th Asian Conference (ACIIDS 2014), 8398, 23–32.
@inproceedings{mohammad2014lengthrange,
keywords = {conference},
focus = {timeseries},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Exact Discovery of Length-Range Motifs},
booktitle = {Intelligent Information and Database Systems --- 6th Asian Conference (ACIIDS 2014)},
series = {Lecture Notes in Computer Science},
volume = {8398},
pages = {23--32},
publisher = {Springer},
address = {Bangkok, Thailand},
year = {2014},
doi = {10.1007/978-3-319-05458-2_3}
}
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2014). Why Should We Imitate Robots? Proceedings of the 13th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2014), 1499–1500. Extended abstract.
@inproceedings{mohammad2014whyimitate,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Why Should We Imitate Robots?},
booktitle = {Proceedings of the 13th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2014)},
pages = {1499--1500},
publisher = {IFAAMAS/ACM},
address = {Paris, France},
year = {2014},
note = {Extended abstract}
}
Learning through imitation is a promising technology for robots that inhabit human’s physical and social spaces. Previous research in HRI have shown that human’s subjective evaluation of robot’s abilities affect the way people interact with robots. Given that one of the major challenges in learning from demonstration (imitation) in robotics is the limited number of training examples that the demonstrator is usually willing to provide, it would be beneficial to design the interaction context in such a way to increase human’s subjective evaluation of the robot’s imitative skills. We propose back imitation as a way to achieve that goal. Two experiments were conducted – involving 36 subjects and 124 sessions– to evaluate the effect of back imitation (and a variant called mutual imitation) on human’s subjective evaluation of the robot along several dimensions including imitation skill, motion human likeness, interaction quality, humanness and likability. The paper reports the results of these experiments and discusses their implications for the design of imitation interactions.
Robotics & HRI
Lala, D., Nishida, T., & Mohammad, Y. F. O. (2014). A Joint Activity Theory Analysis of Body Interactions in Multiplayer Virtual Basketball. Proceedings of the 28th International BCS Human Computer Interaction Conference (BCS-HCI 2014).
@inproceedings{lala2014jointactivity,
keywords = {conference},
focus = {robotics},
author = {Lala, Divesh and Nishida, Toyoaki and Mohammad, Yasser F. O.},
title = {A Joint Activity Theory Analysis of Body Interactions in Multiplayer Virtual Basketball},
booktitle = {Proceedings of the 28th International BCS Human Computer Interaction Conference (BCS-HCI 2014)},
series = {Workshops in Computing},
publisher = {British Computer Society},
address = {Southport, UK},
year = {2014}
}
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2014). Human-Like Motion of a Humanoid in a Shadowing Task. International Conference on Collaboration Technologies and Systems (CTS 2014), Workshop on Collaborative Robots and Human Robot Interaction (CR-HRI), 123–130.
@inproceedings{mohammad2014shadowing,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Human-Like Motion of a Humanoid in a Shadowing Task},
booktitle = {International Conference on Collaboration Technologies and Systems (CTS 2014), Workshop on Collaborative Robots and Human Robot Interaction (CR-HRI)},
pages = {123--130},
publisher = {IEEE},
address = {Minneapolis, MN, USA},
year = {2014},
doi = {10.1109/CTS.2014.6867553}
}
Humanoid robots have –by definition– some level of human-likeness in body form. According to previous research in HRI, this leads to a higher expectation of human-like behavior. Nevertheless, human-likeness is not an easy notion to define for motion even in a task as straight forward as real-time motion copying (the shadowing task) as this paper will try to argue. The main hypothesis of this paper is that subjective evaluation of robot’s motion’s human-likeness depends not only on the objective similarity between robot’s motion and human motion but also on the interaction context (e.g. whether or not the human have previously engaged in mutual or back imitation with the robot). Moreover, the paper proposes two features of motion similarity that affect subjective evaluation of human-likeness and accuracy in the shadowing task and shows that human-likeness is a different attribution dimension form both accuracy and humanness (measured using human-nature traits). The paper reports a controlled user study involving 36 participants and 108 HRI sessions to support these claims.
ML & Time-Series
Mohammad, Y. F. O., & Nishida, T. (2014). Scale Invariant Multi-Length Motif Discovery. Modern Advances in Applied Intelligence — 27th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2014), 8482, 417–426.
@inproceedings{mohammad2014scaleinvariant,
keywords = {conference},
focus = {timeseries},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Scale Invariant Multi-Length Motif Discovery},
booktitle = {Modern Advances in Applied Intelligence --- 27th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2014)},
series = {Lecture Notes in Computer Science},
volume = {8482},
pages = {417--426},
publisher = {Springer},
address = {Kaohsiung, Taiwan},
year = {2014},
doi = {10.1007/978-3-319-07467-2_44}
}
Discovering approximately recurrent motifs (ARMs) in timeseries is an active area of research in data mining. Exact motif discovery was later defined as the problem of efficiently finding the most similar pairs of timeseries subsequences and can be used as a basis for discovering ARMs. The most efficient algorithm for solving this problem is the MK algorithm which was designed to find a single pair of timeseries subsequences with maximum similarity at a known length. Available exact solutions to the problem of finding top K similar subsequence pairs at multiple lengths (which can be the basis of ARM discovery) are not scale invariant. This paper proposes a new algorithm for solving this problem efficiently using scale invariant distance functions and applies it to both real and synthetic dataset.
Robotics & HRI
Tatsumi, S., Mohammad, Y. F. O., Ohmoto, Y., & Nishida, T. (2014). Detection of Hidden Laughter for Human-Agent Interaction. 18th International Conference on Knowledge-Based and Intelligent Information and Engineering Systems (KES 2014), 35, 1053–1062.
@inproceedings{tatsumi2014laughter,
keywords = {conference},
focus = {robotics},
author = {Tatsumi, Shiho and Mohammad, Yasser F. O. and Ohmoto, Yoshimasa and Nishida, Toyoaki},
title = {Detection of Hidden Laughter for Human-Agent Interaction},
booktitle = {18th International Conference on Knowledge-Based and Intelligent Information and Engineering Systems (KES 2014)},
series = {Procedia Computer Science},
volume = {35},
pages = {1053--1062},
publisher = {Elsevier},
address = {Gdynia, Poland},
year = {2014},
doi = {10.1016/j.procs.2014.08.192}
}
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2014). Effect of Back and Mutual Imitation on Human’s Perception of a Humanoid’s Imitative Skill. The 23rd IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN 2014), 788–795.
@inproceedings{mohammad2014backmutual,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Effect of Back and Mutual Imitation on Human's Perception of a Humanoid's Imitative Skill},
booktitle = {The 23rd IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN 2014)},
pages = {788--795},
publisher = {IEEE},
address = {Edinburgh, UK},
year = {2014},
doi = {10.1109/ROMAN.2014.6926349}
}
A promising technology for teaching robots new skills is imitation learning. For robots to learn through from demonstrations, humans should be motivated to teach them by providing the required demonstrations and feedback for robot’s trials at imitation. HRI research has shown that human’s perception of robot’s skills affect robot’s acceptability and human willingness to interact with it. Even though there is currently large literature in imitation learning and learning from demonstrations (LfD), little attention was given to studying methods to improve humans’ perception of robot’s imitative skill (other than objectively improving these skills). In this paper, we study the effect of back and simple mutual imitation on human’s perception of robot’s imitative skill and argue that both of these simple manipulations of the interaction improve human’s subjective evaluation of the robot’s imitative skill in terms of accuracy and overall performance while having no effect on subjective evaluation of naturalness and speed.
Robotics & HRI
Mohammad, Y., & Nishida, T. (2014). Robust Learning from Demonstrations Using Multidimensional SAX. 14th International Conference on Control, Automation and Systems (ICCAS 2014), 64–71. Outstanding Paper Award.
ICCAS 2014 Outstanding Paper
@inproceedings{mohammad2014robustlfd,
award = {ICCAS 2014 Outstanding Paper},
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser and Nishida, Toyoaki},
title = {Robust Learning from Demonstrations Using Multidimensional {SAX}},
booktitle = {14th International Conference on Control, Automation and Systems (ICCAS 2014)},
pages = {64--71},
publisher = {IEEE},
address = {Seoul, South Korea},
year = {2014},
note = {Outstanding Paper Award}
}
Learning from demonstrations (LfD) is gaining more popularity in robotics due to its promise of providing a human-friendly technique for teaching robots new skills by robotics-naive users. The two main approaches to LfD are dynamic motor primitives (DMP) which models demonstrated motions as dynamical systems with the advantage flexibility in changing the motion’s starting position, goal or speed and Gaussian Mixture Modelling/ Gaussian Mixture Regression (GMM/GMR) which represents demonstrated motions as mixtures of Gaussians with the advantage of keeping track of the correlations between different dimensions of learned motions and automatic extraction of motion variability along these dimensions. This paper introduces a third approach that relies on symbolization of demonstrated motions by extending the Symbolic Aggregate approXimation (SAX) to handle multiple dimensions of data. The proposed approach is shown through several synthetic and real-data evaluations to be more resistant to bursts of noise that usually appear in motion capture data used to represent the demonstrations. The paper also discusses possible ways to combine SAX based LfD with DMP and GMM/GMR in order to preserve the advantages of these two approaches while providing superior burst noise resistance.
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2013). Tackling the Correspondence Problem — Closed-Form Solution for Gesture Imitation by a Humanoid’s Upper Body. Active Media Technology — 9th International Conference (AMT 2013), 8210, 84–95.
@inproceedings{mohammad2013correspondence,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Tackling the Correspondence Problem --- Closed-Form Solution for Gesture Imitation by a Humanoid's Upper Body},
booktitle = {Active Media Technology --- 9th International Conference (AMT 2013)},
series = {Lecture Notes in Computer Science},
volume = {8210},
pages = {84--95},
publisher = {Springer},
address = {Maebashi, Japan},
year = {2013},
doi = {10.1007/978-3-319-02750-0_9}
}
ML & Time-Series
Mohammad, Y. F. O., & Nishida, T. (2013). Approximately Recurring Motif Discovery Using Shift Density Estimation. Recent Trends in Applied Artificial Intelligence — 26th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2013), 7906, 141–150.
@inproceedings{mohammad2013armd,
keywords = {conference},
focus = {timeseries},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Approximately Recurring Motif Discovery Using Shift Density Estimation},
booktitle = {Recent Trends in Applied Artificial Intelligence --- 26th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2013)},
series = {Lecture Notes in Computer Science},
volume = {7906},
pages = {141--150},
publisher = {Springer},
address = {Amsterdam, The Netherlands},
year = {2013},
doi = {10.1007/978-3-642-38577-3_15}
}
Approximately Recurring Motif (ARM) discovery is the problem of finding unknown patterns that appear frequently in real valued timeseries. In this paper, we propose a novel algorithm for solving this problem that can achieve performance comparable with the most accurate algorithms to solve this problem with a speed comparable to the fastest ones. The main idea behind the proposed algorithm is to convert the problem of ARM discovery into a density estimation problem in the single dimensionality shift-space (rather than in the original time-series space). This makes the algorithm more robust to short noise bursts that can dramatically affect the performance of most available algorithms. The paper also reports the results of applying the proposed algorithm to synthetic and real-world datasets.
Robotics & HRI
Mohammad, Y. F. O., Nishida, T., & Nakazawa, A. (2013). Arm Pose Copying for Humanoid Robots. IEEE International Conference on Robotics and Biomimetics (ROBIO 2013), 897–904.
@inproceedings{mohammad2013armpose,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki and Nakazawa, Atsushi},
title = {Arm Pose Copying for Humanoid Robots},
booktitle = {IEEE International Conference on Robotics and Biomimetics (ROBIO 2013)},
pages = {897--904},
publisher = {IEEE},
address = {Shenzhen, China},
year = {2013},
doi = {10.1109/ROBIO.2013.6739576}
}
Learning by imitation is becoming increasingly important for teaching humanoid robots new skills. The simplest form of imitation is behavior copying in which the robot is minimizing the difference between its perceived motion and that of the imitated agent. One problem that must be solved even in this simplest of all imitation tasks is calculating the learner’s pose corresponding to the perceived pose of the agent it is imitating. This paper presents a general framework for solving this problem in closed form for the arms of a generalized humanoid robot of which most available humanoids are special cases. The paper also reports the evaluation of the proposed system for real and simulated robots.
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2013). Learning Sensorimotor Concepts Without Reinforcement. Lifelong Machine Learning — 2013 AAAI Spring Symposium, SS-13-05.
@inproceedings{mohammad2013sensorimotor,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Learning Sensorimotor Concepts Without Reinforcement},
booktitle = {Lifelong Machine Learning --- 2013 AAAI Spring Symposium},
series = {AAAI Technical Report},
volume = {SS-13-05},
publisher = {AAAI},
address = {Palo Alto, CA, USA},
year = {2013}
}
Agents engaged in lifelong learning can benefit from the ability to acquire new concepts from continuous interaction with objects in their environments which is a ubiquitous ability in humans. This paper advocates the use of sensorimotor concepts that combine perceptual and actuation patterns. Related representations to sensorimotor concepts are Predictive State Representation in dynamical systems, Affordance Based Concepts in language and Skills in reinforcement learning. The paper proposes a system for learning generalized sensorimotor concepts from unsegmented interactions between the agent and the objects in its environment that works in continuous action and observation spaces and in the same time require no reinforcement signals. A proof-of-concept experiment with the proposed system on a simulated e-puck robot is reported to support the applicability of the proposed approach.
Robotics & HRI
Lala, D., Nishida, T., & Mohammad, Y. (2013). Unsupervised Gesture Recognition System for Learning Manipulative Actions in Virtual Basketball. Proceedings of the 1st International Conference on Human-Agent Interaction (IHAI 2013).
@inproceedings{lala2013gesture,
keywords = {conference},
focus = {robotics},
author = {Lala, Divesh and Nishida, Toyoaki and Mohammad, Yasser},
title = {Unsupervised Gesture Recognition System for Learning Manipulative Actions in Virtual Basketball},
booktitle = {Proceedings of the 1st International Conference on Human-Agent Interaction (iHAI 2013)},
year = {2013}
}
Robotics & HRI
Mohammad, Y. F. O., Ohmoto, Y., & Nishida, T. (2012). Common Sensorimotor Representation for Self-Initiated Imitation Learning. Advanced Research in Applied Artificial Intelligence — 25th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2012), 7345, 381–390.
@inproceedings{mohammad2012commonsensorimotor,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Ohmoto, Yoshimasa and Nishida, Toyoaki},
title = {Common Sensorimotor Representation for Self-Initiated Imitation Learning},
booktitle = {Advanced Research in Applied Artificial Intelligence --- 25th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2012)},
series = {Lecture Notes in Computer Science},
volume = {7345},
pages = {381--390},
publisher = {Springer},
address = {Dalian, China},
year = {2012},
doi = {10.1007/978-3-642-31087-4_40}
}
Internal representation is an important design decision in any imitation learning system. Actions and perceptual spaces where separate in classical AI due to standard sense-process-act loop. Recently another representation that combines the two spaces into what we call a common sensorimotor space was inspired by the discovery of mirror neurons in animals and humans. The justification of this move is usually biological plausibility. This paper reports a series of experiments comparing these two alternatives for self-initiated imitation tasks. The results of these experiments show that using a common sensorimotor representation allows the system to achieve higher accuracy and sensitivity. This is shown to be true (for our scenarios) even when the dimensionality of the common sensorimotor representation is higher than the dimensionality of the separate perceptual space. It also allows for an easier behavior generation mechanism and ensures reproducibility of learned behavior by the learner.
Mohammad, Y., & Nishida, T. (2012). Unsupervised Discovery of Basic Human Actions from Activity Recording Datasets. IEEE/SICE International Symposium on System Integration (SII 2012), 402–409.
@inproceedings{mohammad2012basichuman,
keywords = {conference},
author = {Mohammad, Yasser and Nishida, Toyoaki},
title = {Unsupervised Discovery of Basic Human Actions from Activity Recording Datasets},
booktitle = {IEEE/SICE International Symposium on System Integration (SII 2012)},
pages = {402--409},
publisher = {IEEE},
address = {Fukuoka, Japan},
year = {2012},
doi = {10.1109/SII.2012.6426960}
}
Human Behavior Understanding (HBU) is a major challenge facing intelligent agents. Most approaches to solve this problem assume a recognition/detection context in which the agent/robot tries to match the perceived behavior to one or more predefined motion patterns (e.g. walking, running etc). A more challenging problem is discovering these motion patterns without apriori assumption about the motions in the data, their duration or their numbers. This paper proposes the utilization of a novel motif discovery algorithm based on the exact MK algorithm to discover basic actions in activity records. The proposed system was evaluated on real records of full body motions and is shown in this paper to achieve high accuracy compared with a recently proposed motif discovery algorithm applied to the same dataset.
Robotics & HRI
Mohammad, Y., & Nishida, T. (2012). Self-Initiated Imitation Learning: Discovering What to Imitate. 12th International Conference on Control, Automation and Systems (ICCAS 2012), 726–732.
@inproceedings{mohammad2012selfinitiated,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser and Nishida, Toyoaki},
title = {Self-Initiated Imitation Learning: Discovering What to Imitate},
booktitle = {12th International Conference on Control, Automation and Systems (ICCAS 2012)},
pages = {726--732},
publisher = {IEEE},
address = {Jeju, South Korea},
year = {2012}
}
Imitation learning is an important area in robotics and agents research because it provides an easy way for robot programming and also a bootstrapping technique for social learning. Available learning by imitation systems implicitly or explicitly assume that the boundaries of the actions to be imitated are set by the demonstrator and that the robot is in some imitation mode during the whole interaction session. A less researched area is self-initiated imitation in which the robot needs to decide for itself what to imitate from another imitatee that may not be actively involved in the demonstration process. In this paper, we propose a self-initiated imitation engine based on combining techniques from time-series analysis and causality discovery. The paper also reports a series of proof of concept experiments using simulated and real robots. These evaluations show that the proposed approach is capable of discovering important patterns of behavior during the interaction session and faithfully reproduces them.
ML & Time-Series
Mohammad, Y. F. O., Ohmoto, Y., & Nishida, T. (2012). CPMD: A Matlab Toolbox for Change Point and Constrained Motif Discovery. Advanced Research in Applied Artificial Intelligence — 25th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2012), 7345, 114–123.
@inproceedings{mohammad2012cpmd,
keywords = {conference},
focus = {timeseries},
author = {Mohammad, Yasser F. O. and Ohmoto, Yoshimasa and Nishida, Toyoaki},
title = {{CPMD}: A Matlab Toolbox for Change Point and Constrained Motif Discovery},
booktitle = {Advanced Research in Applied Artificial Intelligence --- 25th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2012)},
series = {Lecture Notes in Computer Science},
volume = {7345},
pages = {114--123},
publisher = {Springer},
address = {Dalian, China},
year = {2012},
doi = {10.1007/978-3-642-31087-4_13}
}
Change Point Discovery (CPD) and Constrained Motif Discovery (CMD) are two essential problems in data mining with applications in many fields including robotics, economics, neuroscience and other fields. In this paper, we show that these two problems are related and report the development of a MATLAB Toolbox (CPMD) that encapsulates several useful algorithms including new variants to solve these two related problems. The Toolbox is then used to study the effect of distance function choice in CPD.
ML & Time-Series
Mohammad, Y. F. O., Ohmoto, Y., & Nishida, T. (2012). G-SteX: Greedy Stem Extension for Free-Length Constrained Motif Discovery. Advanced Research in Applied Artificial Intelligence — 25th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2012), 7345, 417–426.
@inproceedings{mohammad2012gstex,
keywords = {conference},
focus = {timeseries},
author = {Mohammad, Yasser F. O. and Ohmoto, Yoshimasa and Nishida, Toyoaki},
title = {{G-SteX}: Greedy Stem Extension for Free-Length Constrained Motif Discovery},
booktitle = {Advanced Research in Applied Artificial Intelligence --- 25th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2012)},
series = {Lecture Notes in Computer Science},
volume = {7345},
pages = {417--426},
publisher = {Springer},
address = {Dalian, China},
year = {2012},
doi = {10.1007/978-3-642-31087-4_44}
}
Most available motif discovery algorithms in real-valued time series find approximately recurring patterns of a known length without any prior information about their locations or shapes. In this paper, a new motif discovery algorithm is proposed that has the advantage of requiring no upper limit on the motif length. The proposed algorithm can discover multiple motifs of multiple lengths at once, and can achieve a better accuracy-speed balance compared with a recently proposed motif discovery algorithm. We then briefly report two successful applications of the proposed algorithm to gesture discovery and robot motion pattern discovery.
ML & Time-Series
Mohammad, Y., & Nishida, T. (2011). Discovering Causal Change Relationships Between Processes in Complex Systems. IEEE/SICE International Symposium on System Integration (SII 2011), 12–17.
@inproceedings{mohammad2011causalchange,
keywords = {conference},
focus = {timeseries},
author = {Mohammad, Yasser and Nishida, Toyoaki},
title = {Discovering Causal Change Relationships Between Processes in Complex Systems},
booktitle = {IEEE/SICE International Symposium on System Integration (SII 2011)},
pages = {12--17},
publisher = {IEEE},
address = {Kyoto, Japan},
year = {2011}
}
Complex systems involve the interaction between many processes that may or may not have causal relations to each other. In such systems, discovering causal relations can provide significant insights into the internals of the system and facilitate fault discovery and recovery procedures. In this paper, we provide a novel causality detection algorithm based on robust singular spectrum transform that combines features of autoregressive modeling and perturbation analysis. The proposed approach was evaluated using both synthetic and real data and was shown to provide superior performance to the standard linear Granger-causality test. It also provides a natural way to detect common causes that may give false positives in other causality tests.
ML & Time-Series
Mohammad, Y., & Nishida, T. (2011). On Comparing SSA-Based Change Point Discovery Algorithms. IEEE/SICE International Symposium on System Integration (SII 2011), 938–945. Best Paper Award (Control).
IEEE SII 2011 Best Paper
@inproceedings{mohammad2011ssacomparison,
award = {IEEE SII 2011 Best Paper},
keywords = {conference},
focus = {timeseries},
author = {Mohammad, Yasser and Nishida, Toyoaki},
title = {On Comparing {SSA}-Based Change Point Discovery Algorithms},
booktitle = {IEEE/SICE International Symposium on System Integration (SII 2011)},
pages = {938--945},
publisher = {IEEE},
address = {Kyoto, Japan},
year = {2011},
note = {Best Paper Award (Control)}
}
Change point discovery is an important problem in data mining and industrial systems. Different approaches have been proposed and some of the most promising approaches are based on singular spectrum analysis (SSA). These algorithms have the advantages of requiring no ad-hoc tuning for different types of signals and having a built-in noise attenuation mechanism. In this paper we try to unify these approaches and present a novel method for comparing change point discovery algorithms. We then use the proposed method to compare different SSA based change point discovery algorithms. Even though we focused on comparing only SSA based algorithms, the proposed metric applicable to any kind of change point discovery algorithm and have the advantages of requiring no localization steps, and being independent of any predefined thresholds (unlike traditional metrics).
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2010). Incremental Gesture Discovery for Interactive Robots. IEEE International Conference on Robotics and Biomimetics (ROBIO 2010), 185–189.
@inproceedings{mohammad2010incremental,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Incremental Gesture Discovery for Interactive Robots},
booktitle = {IEEE International Conference on Robotics and Biomimetics (ROBIO 2010)},
pages = {185--189},
publisher = {IEEE},
address = {Tianjin, China},
year = {2010},
doi = {10.1109/ROBIO.2010.5723324}
}
An incremental method for gesture discovery from continuous time streams based on CMD algorithm is introduced and evaluated. Results show that the method is as accurate as batch discovery and that utilizing action streams is effective in removing false negatives
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2010). Learning Interaction Protocols Using Augmented Bayesian Networks Applied to Guided Navigation. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2010), 4119–4126.
@inproceedings{mohammad2010augmentedbayes,
keywords = {conference, important},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Learning Interaction Protocols Using Augmented Bayesian Networks Applied to Guided Navigation},
booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2010)},
pages = {4119--4126},
publisher = {IEEE},
address = {Taipei, Taiwan},
year = {2010},
doi = {10.1109/IROS.2010.5651719}
}
Research in robot navigation usually concentrates on implementing navigation algorithms that allow the robot to navigate without human aid. In many real world situations, it is desirable that the robot is able to understand natural gestures from its user or partner and use this understanding to guide its navigation. Some algorithms already exist for learning natural gestures and/or their associated actions but most of these systems does not allow the robot to automatically generate the associated controller that allows it to actually navigate in the real environment. Furthermore, a technique is needed to combine the gestures/actions learned from interacting with multiple users or partners. This paper resolves these two issues and provides a complete system that allows the robot to learn interaction protocols and act upon them using only unsupervised learning techniques and enables it to combine the protocols learned from multiple users/partners. The proposed approach is general and can be applied to other interactive tasks as well. This paper also provides a real world experiment involving 18 subjects and 72 sessions that supports the ability of the proposed system to learn the needed gestures and to improve its knowledge of different gestures and their associations to actions over time.
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2010). Learning Spontaneous Nonverbal Behavior Using a Three Layers Hierarchy. Proceedings of the 10th WSEAS International Conference on Applied Computer Science, 430–435.
@inproceedings{mohammad2010threelayers,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Learning Spontaneous Nonverbal Behavior Using a Three Layers Hierarchy},
booktitle = {Proceedings of the 10th WSEAS International Conference on Applied Computer Science},
pages = {430--435},
address = {Iwate, Japan},
year = {2010}
}
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2010). Down-Up-Down Behavior Generation for Interactive Robots. Trends in Applied Intelligent Systems — 23rd International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2010), 6096, 92–101.
@inproceedings{mohammad2010downupdown,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Down-Up-Down Behavior Generation for Interactive Robots},
booktitle = {Trends in Applied Intelligent Systems --- 23rd International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2010)},
series = {Lecture Notes in Computer Science},
volume = {6096},
pages = {92--101},
publisher = {Springer},
address = {Cordoba, Spain},
year = {2010},
doi = {10.1007/978-3-642-13022-9_10}
}
Behavior generation in humans and animals usually employs a combination of bottom-up and top-down patterns. Most available robotic architectures utilize either bottom-up or top-down activation including hybrid architectures. In this paper, we propose a behavior generation mechanism that can seamlessly combine these two strategies. One of the main advantages of the proposed approach is that it can naturally combine both bottom-up and top-down behavior generation mechanisms which can produce more natural behavior. This is achieved by utilizing results from the theory of simulation in neuroscience which tries to model the mechanism used in human infants to develop a theory of mind. The proposed approach was tested in modeling spontaneous gaze control during natural face to face interactions and provided more natural, human-like behavior compared with a state-of-the-art gaze controller that utilized a bottom-up approach.
Robotics & HRI
Mohammad, Y. F. O., Nishida, T., & Okada, S. (2009). Unsupervised Simultaneous Learning of Gestures, Actions and Their Associations for Human-Robot Interaction. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2009), 2537–2544.
@inproceedings{mohammad2009simultaneous,
keywords = {conference, important},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki and Okada, Shogo},
title = {Unsupervised Simultaneous Learning of Gestures, Actions and Their Associations for Human-Robot Interaction},
booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2009)},
pages = {2537--2544},
publisher = {IEEE},
address = {St. Louis, MO, USA},
year = {2009},
doi = {10.1109/IROS.2009.5353987}
}
To enable free natural communication between a human operator and a robot three problems must be faced: Firstly the robot have to know the actions it can do in the world. Secondly the robot must be able to learn the patterns in the perceived behavior of its operator that correspond to commands. Finally the robot needs to know when to execute a specific action based on its perception of the operator’s behavior. In this paper we are interested in free hand gestures as the commanding channel. The most restrictive solution to the aforementioned three problems is to fix the action space (pre-programmed actions), fix the command space (predefined gestures), and fix action-command relation (fixed gesture meanings).Learning by demonstration can be viewed as a technique to relax the first restriction by learning the action space. Gesture interpretation can be viewed as a technique to relax the second restriction by learning the command space. Reinforcement learning can be viewed as a technique for relaxing the third restriction by learning action-command associations (policy). In this paper we propose a novel technique that allows the robot to solve these three problems together learning the action space, the command space, and their relations by just another robot operated by a human operator. The main technical contribution of this paper is the introduction of a novel algorithm that allows the robot to segment and discover patterns in its perceived signals without any prior knowledge of the number of different patterns, their occurrences or lengths. The second contribution is using a Ganger-Causality based test to limit the search space for actions and commands utilizing their relations and taking into account the autonomy level of the robot. The paper also presents a feasibility study in which the learning robot was able to predict actor’s behavior with 95.2 after monitoring a single interaction between a novice operator and a WOZ operated robot representing the actor.
ML & Time-Series
Mohammad, Y. F. O., & Nishida, T. (2009). Robust Singular Spectrum Transform. Next-Generation Applied Intelligence — 22nd International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2009), 5579, 123–132. Best Paper Award.
IEA/AIE 2009 Best Paper
@inproceedings{mohammad2009rsst,
award = {IEA/AIE 2009 Best Paper},
keywords = {conference},
focus = {timeseries},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Robust Singular Spectrum Transform},
booktitle = {Next-Generation Applied Intelligence --- 22nd International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2009)},
series = {Lecture Notes in Computer Science},
volume = {5579},
pages = {123--132},
publisher = {Springer},
address = {Tainan, Taiwan},
year = {2009},
doi = {10.1007/978-3-642-02568-6_13},
note = {Best Paper Award}
}
Change Point Discovery is a basic algorithm needed in many time series mining applications including rule discovery, motif discovery, casual analysis, etc. Several techniques for change point discovery have been suggested including wavelet analysis, cosine transforms, CUMSUM, and Singular Spectrum Transform. Of these methods Singular Spectrum Transform (SST) have received much attention because of its generality and because it does not require ad-hoc adjustment for every time series. In this paper we show that traditional SST suffers from two major problems: the need to specify five parameters and the rapid reduction in the specificity with increased noise levels. In this paper we define the Robust Singular Spectrum Transform (RSST) that alleviates both of these problems and compare it to RSST using different synthetic and real-world data series.
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2009). Measuring Naturalness During Close Encounters Using Physiological Signal Processing. Next-Generation Applied Intelligence — 22nd International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2009), 5579, 281–290.
@inproceedings{mohammad2009naturalness,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Measuring Naturalness During Close Encounters Using Physiological Signal Processing},
booktitle = {Next-Generation Applied Intelligence --- 22nd International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2009)},
series = {Lecture Notes in Computer Science},
volume = {5579},
pages = {281--290},
publisher = {Springer},
address = {Tainan, Taiwan},
year = {2009},
doi = {10.1007/978-3-642-02568-6_29}
}
Many researchers in the HRI and ECA domains try to build robots and agents that exhibit human-like behavior in real-world close encounter situations. One major requirement for comparing such robots and agents is to have an objective quantitative metric for measuring naturalness in various kinds of interactions. Some researchers have already suggested techniques for measuring stress level, awareness etc using physiological signals like GSR and BVP. One problem of available techniques is that they are only tested with extreme situations and cannot according to the analysis provided in this paper distinguish the response of human subjects in natural interaction situations. One other problem of the available techniques is that most of them require calibration and some times ad-hoc adjustment for every subject. This paper explores the usefulness of various kinds of physiological signals and statistics in distinguishing natural and unnatural partner behavior in a close encounter situation. The paper also explores the usefulness of these statistics in various time slots of the interaction. Based on this analysis a regressor was designed to measure naturalness in close encounter situations and was evaluated using human-human and human-robot interactions and shown to achieve 92.5 situations.
ML & Time-Series
Mohammad, Y., & Nishida, T. (2009). Change Point Detection Using Robust Singular Spectrum Transform Applied to Mining Human-Human Interaction Records. International Symposium on Data Mining and Statistical Science.
@inproceedings{mohammad2009changepoint,
keywords = {conference},
focus = {timeseries},
author = {Mohammad, Yasser and Nishida, Toyoaki},
title = {Change Point Detection Using Robust Singular Spectrum Transform Applied to Mining Human-Human Interaction Records},
booktitle = {International Symposium on Data Mining and Statistical Science},
address = {Japan},
year = {2009}
}
Change Point Discovery is a basic algorithm needed in many time series mining applications including rule discovery, motif discovery, casual analysis, etc. Several techniques for change point discovery have been suggested including wavelet analysis, cosine transforms, CUMSUM, and Singular Spectrum Transform. Of these methods Singular Spectrum Transform (SST) have received much attention because of its generality and because it does not require ad-hoc adjustment for every time series. In this paper we show that traditional SST suffers from two major problems: 1. The need to specify five parameters some of which are very hard to select even if domain knowledge is available. 2. The specificity of the transform degrades rapidly with increased noise levels specially when the background signal of the time series is zero which is true for many human generated signals. The main contributions of this paper are: 1. Defining the RSST transform which requires the specification of only two parameters that can be decided easily based on domain knowledge or visualization, and achieves better performance than traditional Singular Spectrum Transform (SST) in noisy environments. The proposed algorithm is compared to SST and analyzed extensively using several synthetic and real world time series. 2. Applying the algorithm to achieve 8 folds speedup in selecting candidate windows for a motif discovery algorithm 3. Using RSST to define an objective physiological measure of interaction ’naturalness’ in human-human explanation scenarios.
Robotics & HRI
Mohammad, Y., Xu, Y., Matsumura, K., & Nishida, T. (2008). The H3R Explanation Corpus: Human-Human and Base Human-Robot Interaction Dataset. The Fourth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP 2008), 201–206.
@inproceedings{mohammad2008h3r,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser and Xu, Yong and Matsumura, Kenichi and Nishida, Toyoaki},
title = {The {H3R} Explanation Corpus: Human-Human and Base Human-Robot Interaction Dataset},
booktitle = {The Fourth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP 2008)},
pages = {201--206},
publisher = {IEEE},
address = {Sydney, Australia},
year = {2008}
}
Natural interaction between humanoid robots and humans is one of the major goals of the HRI field. Two major requirements for advancing this direction of research are the availability of human–human and base human–robot interaction datasets for training and evaluation purposes and the availability of general agreed upon objective metrics for judging the performance of proposed robots and algorithms. In this paper we report details of the H 3R Explanation Corpus dataset of human–human and base human–robot interactions in assembly/disassembly explanation scenarios that combines five kinds of data: video, audio, motion tracking, subjective, and physiological data. 44 subjects and 66 sessions were conducted during this experiment. The corpus contains 22 natural Human–Human interactions, 22 un-natural Human–Human interactions, and 22 baseline Human-Robot interactions. To our best knowledge this is the first database that combines these five data types and three types of interactions. The paper also reports the first usage of this explanation corpus to compare subjective and physiological evaluations of various dimensions of listener’s behavior.
ML & Time-Series
Mohammad, Y., & Nishida, T. (2008). Constrained Motif Discovery. International Symposium on Data Mining and Statistical Science (DMSS 2008), 16–19.
@inproceedings{mohammad2008dmss,
keywords = {conference},
focus = {timeseries},
author = {Mohammad, Yasser and Nishida, Toyoaki},
title = {Constrained Motif Discovery},
booktitle = {International Symposium on Data Mining and Statistical Science (DMSS 2008)},
pages = {16--19},
address = {Tokyo, Japan},
year = {2008}
}
The goal of motif discovery algorithms is to efficiently find unknown recurring patterns in time series. Most available algorithms cannot utilize domain knowledge in any way which results in quadratic or at least super-linear time and space complexity. In this paper we define the Constrained Motif Discovery problem which enables utilization of domain knowledge into the motif discovery process. The paper then provides two algorithms called MCFull and MCInc for efficiently solving the constrained motif discovery problem. We also show that most unconstrained motif discovery problems be converted into constrained ones using a change-point detection algorithm. A novel change-point detection algorithm called the Robust Singular Spectrum Transform (RSST) is then introduced and compared to traditional Singular Spectrum Transform using synthetic and real-world data sets. The results show that RSST achieves higher specificity and is more adequate for finding constraints to convert unconstrained motif discovery problems to constrained ones that can be solved using MCFull and MCInc. We then compare the combination of RSST and MCFull or MCInc with two state-of-the-art motif discovery algorithms on a large set of synthetic time series. The results show that the proposed algorithms provided four to ten folds increase in speed compared the unconstrained motif discovery algorithms studied without any loss of accuracy. One of the proposed constrained motif discovery algorithms (MCFull) is then used along with the RSST algorithm in a real world human-robot interaction experiment to enable the robot to learn free hand gestures, actions, and their associations by humans and other robots interacting.
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2008). Human Adaptation to a Miniature Robot: Precursors of Mutual Adaptation. The 17th IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN 2008), 124–129.
@inproceedings{mohammad2008adaptation,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Human Adaptation to a Miniature Robot: Precursors of Mutual Adaptation},
booktitle = {The 17th IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN 2008)},
pages = {124--129},
publisher = {IEEE},
address = {Munich, Germany},
year = {2008},
doi = {10.1109/ROMAN.2008.4600654}
}
Mutual adaptation is an important phenomenon in human-human communications. Traditionally HRI research were more interested in investigating adaptation of the robot to the human using machine learning techniques but the possibility of utilizing the natural ability of humans to adapt to other humans and artifacts including robots is recently becoming more attractive. This paper presents some of the results from an experiment conducted to investigate the interaction patterns and effectiveness of motion cues as a feedback modality between a human operator and a miniature robot in a confined collaborative navigation task. The results presented in this paper show evidence of human adaptation to the robot and moreover suggest that the adaptation rate is not constant or continuous in time but is discontinuous and nonlinear. The results also show evidence of a starting stage before the adaptation with duration dependent on the expectations of the human regarding the capabilities of the robot in the given task. The paper investigates how to utilize these and related findings for building robots capable of not only adapting to human operators but can also help those operators adapt to them.
Robotics & HRI
Mohammad, Y., & Nishida, T. (2008). Getting Feedback from a Miniature Robot. IEEE International Conference on Information and Automation (ICIA 2008), 941–947.
@inproceedings{mohammad2008gettingfeedback,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser and Nishida, Toyoaki},
title = {Getting Feedback from a Miniature Robot},
booktitle = {IEEE International Conference on Information and Automation (ICIA 2008)},
pages = {941--947},
publisher = {IEEE},
address = {Changsha, China},
year = {2008}
}
The HRI field of research has gained much attention recently because of the expected importance of well designed interaction modalities with social robots. To achieve natural interaction between the robot and the human, a feedback mechanism from the robot to the human needs to be designed that allows the robot to express its internal state to the human in a natural way. Verbal and nonverbal feedback from humanoid robots or humanoid robotic heads have been widely studied but there is little comparable research about the possible feedback mechanisms of non-humanoid and especially miniature robots. In this paper a comparison between using verbal feedback and motion cues is conducted. The results of the experiment showed that there is no significant difference in the task completion accuracy and time or in the feeling of naturalness between these two modalities and there is a statistically significant improvement when using any of them compared with the no-feedback (control) case. Moreover the subjects selected the motion cues feedback mechanism more frequently as the preferred feedback modality for them.
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2008). Reactive Gaze Control for Natural Human-Robot Interactions. IEEE Conference on Robotics, Automation and Mechatronics (RAM 2008), 47–54.
@inproceedings{mohammad2008reactivegaze,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Reactive Gaze Control for Natural Human-Robot Interactions},
booktitle = {IEEE Conference on Robotics, Automation and Mechatronics (RAM 2008)},
pages = {47--54},
publisher = {IEEE},
address = {Chengdu, China},
year = {2008},
doi = {10.1109/RAMECH.2008.4681417}
}
Nonverbal behavior during human-human close encounters is critical to the accomplishment of natural interaction. For this reason, humanoid robots trying to achieve natural interactions with humans should be able to understand and synthesis nonverbal behavior in a way that mimics the human use of it. One of the most important situations during natural human-robot interactions is the explanation scenario in which the human is explaining a task to the robot using natural verbal and nonverbal behavior. This situation occurs frequently in many HRI applications and is critical to the success of the Robots as Knowledge Media project suggested by the authors. In this paper the implementation of a humanoid robot that can show human like gaze control during explanation settings based only on reactive processing is presented. The software of the robot is based on the EICA architecture designed to combine autonomy with interactivity in the lowest level of the system. The details of the implementation and analysis of the naturalness of behavior and the effect of noisy input is presented in this paper.
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2008). A Cross-Platform Robotic Architecture for Autonomous Interactive Robots. New Frontiers in Applied Artificial Intelligence — 21st International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2008), 5027, 108–118. Best Paper Award.
IEA/AIE 2008 Best Paper
@inproceedings{mohammad2008crossplatform,
award = {IEA/AIE 2008 Best Paper},
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {A Cross-Platform Robotic Architecture for Autonomous Interactive Robots},
booktitle = {New Frontiers in Applied Artificial Intelligence --- 21st International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2008)},
series = {Lecture Notes in Computer Science},
volume = {5027},
pages = {108--118},
publisher = {Springer},
address = {Wroclaw, Poland},
year = {2008},
doi = {10.1007/978-3-540-69052-8_12},
note = {Best Paper Award}
}
HRI is a growing field of research that targets the development of robots which are easy to operate, more engaging and more entertaining. Most robotic architectures available including reactive, deliberative, and hybrid architectures were developed with autonomous robots in mind. Although robots developed for HRI are usually autonomous, their autonomy is combined with interactivity which adds more challenges on the design of the robotic architectures supporting them. Among these challenges are the flexibility in combining deliberative and interactive processes, low level attention focusing, and flexible action integration. This paper reports the lowest level of specification of a new cross-platform robotic architecture for HRI applications called EICA (The Embodied Interactive Control Architecture) that was designed to meet those challenges emphasizing how low level attention focusing and action integration are implemented. Two of the main advantages of the proposed architecture are real-time architectural modifiability and supporting adaptability and learning through interactions between processes. The paper describes in details the L 0EICA based design of two gaze controllers that uses the aforementioned properties of the architecture to achieve human-like nonverbal behavior during explanation scenarios. Robotic Architectures Action Integration HRI Gaze Control
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2008). Two Layers Action Integration for HRI — Action Integration with Attention Focusing for Interactive Robots. Proceedings of the Fifth International Conference on Informatics in Control, Automation and Robotics (ICINCO 2008), 41–48.
@inproceedings{mohammad2008twolayers,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Two Layers Action Integration for {HRI} --- Action Integration with Attention Focusing for Interactive Robots},
booktitle = {Proceedings of the Fifth International Conference on Informatics in Control, Automation and Robotics (ICINCO 2008)},
pages = {41--48},
publisher = {INSTICC Press},
address = {Funchal, Madeira, Portugal},
year = {2008}
}
Robotics & HRI
Mohammad, Y., & Nishida, T. (2008). Natural Listening for a Humanoid Robot. Proceedings of the International Conference on Informatics Education and Research for Knowledge-Circulating Society (ICKS 2008), 153–156.
@inproceedings{mohammad2008naturallistening,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser and Nishida, Toyoaki},
title = {Natural Listening for a Humanoid Robot},
booktitle = {Proceedings of the International Conference on Informatics Education and Research for Knowledge-Circulating Society (ICKS 2008)},
pages = {153--156},
publisher = {IEEE},
address = {Kyoto, Japan},
year = {2008}
}
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2007). TalkBack: Feedback from a Miniature Robot. AI 2007: Advances in Artificial Intelligence — 20th Australian Joint Conference on Artificial Intelligence, 4830, 357–366.
@inproceedings{mohammad2007talkback,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {{TalkBack}: Feedback from a Miniature Robot},
booktitle = {AI 2007: Advances in Artificial Intelligence --- 20th Australian Joint Conference on Artificial Intelligence},
series = {Lecture Notes in Computer Science},
volume = {4830},
pages = {357--366},
publisher = {Springer},
address = {Gold Coast, Australia},
year = {2007},
doi = {10.1007/978-3-540-76928-6_37}
}
A prerequisite of any successful social robot is the ability to express its internal state and intention to humans in a natural way. Many researchers studied verbal and nonverbal feedback from humanoid robots or humanoid robotic heads but there is little research about the possible feedback mechanisms of non-humanoid and especially miniature robots. The TalkBack experiment is a trial to fill this gap by investigating the effectiveness of using motion cues as a feedback mechanism and comparing it to verbal feedback. The results of the experiment showed that there is no significant difference in the task completion accuracy and time or in the feeling of naturalness between these two modalities and there is a statistically significant improvement when using any of them compared with the no-feedback case. Moreover the subjects selected the motion cues feedback mechanism more frequently as the preferred feedback modality for them.
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2007). Intention Through Interaction: Toward Mutual Intention in Real World Interactions. New Trends in Applied Artificial Intelligence — 20th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2007), 4570, 115–125.
@inproceedings{mohammad2007intention,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Intention Through Interaction: Toward Mutual Intention in Real World Interactions},
booktitle = {New Trends in Applied Artificial Intelligence --- 20th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE 2007)},
series = {Lecture Notes in Computer Science},
volume = {4570},
pages = {115--125},
publisher = {Springer},
address = {Kyoto, Japan},
year = {2007},
doi = {10.1007/978-3-540-73325-6_12}
}
Human-Artifact interaction in real world situations is currently an active area of research due to the importance foreseen of the social capabilities of near future robots and other intelligent artifacts in integrating them into the human society. In this paper a new paradigm for mutual intention in human-artifact interactions based on the embodied computing paradigm called is introduced with theoretical analysis of its relation to the embodiment framework. As examples of the practical use of the framework to replace traditional symbolic based intention understanding systems, the authors’ preliminary work in a real-world agent architecture (IECA) and a natural drawing environment (NaturalDraw) is briefed.
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2007). NaturalDraw: Interactive Perception Based Drawing for Everyone. Proceedings of the 12th International Conference on Intelligent User Interfaces (IUI 2007), 251–260.
@inproceedings{mohammad2007naturaldraw,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {{NaturalDraw}: Interactive Perception Based Drawing for Everyone},
booktitle = {Proceedings of the 12th International Conference on Intelligent User Interfaces (IUI 2007)},
pages = {251--260},
publisher = {ACM},
address = {Honolulu, HI, USA},
year = {2007},
doi = {10.1145/1216295.1216340}
}
Robotics & HRI
Mohammad, Y., Ohya, T., Hiramatsu, T., Sumi, Y., & Nishida, T. (2007). Embodiment of Knowledge into the Interaction and Physical Domains Using Robots. International Conference on Control, Automation and Systems (ICCAS 2007), 737–744.
@inproceedings{mohammad2007embodiment,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser and Ohya, Taku and Hiramatsu, Tatsuya and Sumi, Yasuyuki and Nishida, Toyoaki},
title = {Embodiment of Knowledge into the Interaction and Physical Domains Using Robots},
booktitle = {International Conference on Control, Automation and Systems (ICCAS 2007)},
pages = {737--744},
address = {Seoul, South Korea},
year = {2007}
}
Robots can do more for the human society than wandering around collecting garbage, entertaining children and playing soccer. In this paper we present an ongoing research toward the realization of the robot as an embodied knowledge media. In this framework, the robot is used to extract knowledge from expert humans through natural interactions, and then to present this knowledge to other humans who need it. The main benefits of using a robot rather than a passive video camera are: 1) The natural interaction between the human expert and the robot will reduce the stress and cognitive load on the human side which allows generation of better videos. 2) The robot will be more effective in extracting useful knowledge by utilizing interaction events to generate embodied knowledge contents in the form of conversation quanta that embody the videos both in the interaction context and in the physical situation. 3) The robot can present knowledge intelligently to other humans who need it, and can generate better views of the knowledge based on the embodied conversation quanta it learned from interacting with the expert and on the natural interaction context and physical context of the current interaction situation. In This paper, the robot as knowledge media theoretical framework will be presented with introduction to the ongoing research conducted in our laboratory toward realizing a humanoid robot that can realize it detailing the past and current contributions and proposing the future research map for this important area of robotic applications. The implementation architecture of the robot based on our proposed EICA system will also be presented with emphasize on the knowledge acquisition behavior of the robot.
Robotics & HRI
Mohammad, Y., Harb, H. M., Mahdy, Y. B., & Mohammed, M. M. (2003). Overcoming Kerberos Structural Limitations. 38th Annual Conference of Statistics, Computer Sciences and Operations Research, 172–187.
@inproceedings{mohammad2003kerberos,
keywords = {conference},
focus = {robotics},
author = {Mohammad, Yasser and Harb, Hany M. and Mahdy, Yousef B. and Mohammed, Montaser M.},
title = {Overcoming Kerberos Structural Limitations},
booktitle = {38th Annual Conference of Statistics, Computer Sciences and Operations Research},
pages = {172--187},
address = {Cairo, Egypt},
year = {2003}
}
International Workshops
Automated Negotiation
Mohammad, Y. (2026). Reinforcement Learning for Automated Negotiation. The 8th Games, Agents, and Incentives Workshop (GAIW@AAMAS 2026).
@inproceedings{mohammad2026rlnegotiation,
keywords = {workshop},
focus = {negotiation},
author = {Mohammad, Yasser},
title = {Reinforcement Learning for Automated Negotiation},
booktitle = {The 8th Games, Agents, and Incentives Workshop (GAIW@AAMAS 2026)},
address = {Paphos, Cyprus},
year = {2026}
}
This tutorial introduces attendees to the problem of building effective negotiation strategies using RL and MARL methods. After providing the motivation for this problem, the tutorial presents the needed theoretical background about automated negotiation and reinforcement learning to appreciate different approaches. Based on this background, the tutorial then introduces a unifying framework and an accompanying open-source library (NegMAS-RL) that can be used to represent most existing research in RL for automated negotiation as well as new solutions not yet attempted. This framework is then used to represent several recent advances in applying RL to automated negotiation and motivate new approaches. The attendees will then learn through a live demonstration (with optional follow-along) how to use this framework to represent, solve and evaluate the solution of a specific problem of automated negotiation in supply chains using the learnt framework.
Automated Negotiation
Mohammad, Y., Nakadai, S., & Greenwald, A. (2024). Automated Negotiation in Supply Chains Game. The First Workshop on Game AI Algorithms and Multi-Agent Learning (GAAMAL@IJCAI 2024).
@inproceedings{mohammad2024scmgame,
keywords = {workshop},
focus = {negotiation},
author = {Mohammad, Yasser and Nakadai, Shinji and Greenwald, Amy},
title = {Automated Negotiation in Supply Chains Game},
booktitle = {The First Workshop on Game AI Algorithms and Multi-Agent Learning (GAAMAL@IJCAI 2024)},
address = {Jeju, South Korea},
year = {2024}
}
Automated Negotiation
Mohammad, Y. (2024). Tentative Acceptance Unique Offers Protocol for Automated Negotiation. The 6th Games, Agents, and Incentives Workshop (GAIW@AAMAS 2024).
@inproceedings{mohammad2024tentative,
keywords = {workshop},
focus = {negotiation},
author = {Mohammad, Yasser},
title = {Tentative Acceptance Unique Offers Protocol for Automated Negotiation},
booktitle = {The 6th Games, Agents, and Incentives Workshop (GAIW@AAMAS 2024)},
address = {Auckland, New Zealand},
year = {2024}
}
Automated Negotiation
Ninagawa, K., Mohammad, Y., & Greenwald, A. (2021). Baseline Strategies for the ANAC Automated Negotiation League. The 3rd Games, Agents, and Incentives Workshop (GAIW@AAMAS 2021).
@inproceedings{ninagawa2021baseline,
keywords = {workshop},
focus = {negotiation},
author = {Ninagawa, Kotone and Mohammad, Yasser and Greenwald, Amy},
title = {Baseline Strategies for the {ANAC} Automated Negotiation League},
booktitle = {The 3rd Games, Agents, and Incentives Workshop (GAIW@AAMAS 2021)},
year = {2021}
}
Automated Negotiation
Mohammad, Y., Nakadai, S., & Greenwald, A. (2019). NegMAS: A Platform for Situated Negotiations. Twelfth International Workshop on Agent-Based Complex Automated Negotiations (ACAN@IJCAI 2019), 57–75.
@inproceedings{mohammad2019negmassituated,
keywords = {workshop},
focus = {negotiation},
author = {Mohammad, Yasser and Nakadai, Shinji and Greenwald, Amy},
title = {{NegMAS}: A Platform for Situated Negotiations},
booktitle = {Twelfth International Workshop on Agent-Based Complex Automated Negotiations (ACAN@IJCAI 2019)},
pages = {57--75},
address = {Macau, China},
year = {2019}
}
Automated negotiations is attracting more interest from researchers in recent years. Most research in this area focuses on negotiation strategy formulation in preset scenarios where the decisions of what to negotiate about, whom to negotiate with, and on which issues are already set to the agent. Moreover, in most cases, the utility functions employed are static and independent between different negotiations. NegMAS (Negotiations Managed by Agent Simulations/ Negotiation Multi-Agent System) was developed to facilitate research and development of agents that negotiate in more dynamic situations characterized by dynamic interrelated utility functions with all negotiation related decisions managed by the agent. This paper introduces NegMAS, its design and overall structure, and evaluates its use in a simple situated negotiations scenario.
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2015). Cultural Difference in Back-Imitation’s Effect on the Perception of Robot’s Imitative Performance. Cultural Robotics — First International Workshop (CR@RO-MAN 2015), 9549, 17–32.
@inproceedings{mohammad2015cultural,
keywords = {workshop},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {Cultural Difference in Back-Imitation's Effect on the Perception of Robot's Imitative Performance},
booktitle = {Cultural Robotics --- First International Workshop (CR@RO-MAN 2015)},
series = {Lecture Notes in Computer Science},
volume = {9549},
pages = {17--32},
publisher = {Springer},
address = {Kobe, Japan},
year = {2015},
doi = {10.1007/978-3-319-42945-8_2}
}
Cultural differences have been documented in different aspects of perception of robots as well as understanding of their behavior. A different line of research in developmental psychology have established a major role of imitation in cultural transfer. This study is a preliminary cross cultural exploration of the effect of imitating the robot (back imitation) on human’s perception of the robot’s imitative skill. In previous research, we have shown that engagement in back imitation with a NAO humanoid robot, results in increased perception of the robot’s imitative skill, motion human likeness, and willingness of future interaction with the robot. This previous work mostly used Japanese university students. In this paper, we report the results of conducting the same study with subjects of two cultures: Japanese and Egyptian university students. The first finding of the study is that the two cultures have widely different expectations of the robot and interaction with it and that some of these differences are significantly reduced after the interaction. The second finding is that Japanese students tended to attribute higher imitation skill and human likeness to the robot they imitated while Egyptian students did not show such tendency. The paper discusses these findings in light of known differences between the two cultures and analyzes the role of expectation in the differences found.
Nishida, T., Nakazawa, A., Ohmoto, Y., Nitschke, C., Mohammad, Y. F. O., Thovutikul, S., Lala, D., Abe, M., & Ookaki, T. (2015). Synthetic Evidential Study as Primordial Soup of Conversation. Databases in Networked Information Systems — 10th International Workshop (DNIS 2015), 8999, 74–83.
@inproceedings{nishida2015primordial,
keywords = {workshop},
author = {Nishida, Toyoaki and Nakazawa, Atsushi and Ohmoto, Yoshimasa and Nitschke, Christian and Mohammad, Yasser F. O. and Thovutikul, Sutasinee and Lala, Divesh and Abe, Masakazu and Ookaki, Takashi},
title = {Synthetic Evidential Study as Primordial Soup of Conversation},
booktitle = {Databases in Networked Information Systems --- 10th International Workshop (DNIS 2015)},
series = {Lecture Notes in Computer Science},
volume = {8999},
pages = {74--83},
publisher = {Springer},
address = {Aizu-Wakamatsu, Japan},
year = {2015},
doi = {10.1007/978-3-319-16313-0_6}
}
Robotics & HRI
Mohammad, Y., & Nishida, T. (2013). What Is Human-Like Motion in a Shadowing Task? IROS 2013 Workshop: Towards Social Humanoid Robots: How to Make Interaction Human-Like?
@inproceedings{mohammad2013humanlike,
keywords = {workshop},
focus = {robotics},
author = {Mohammad, Yasser and Nishida, Toyoaki},
title = {What Is Human-Like Motion in a Shadowing Task?},
booktitle = {IROS 2013 Workshop: Towards Social Humanoid Robots: How to Make Interaction Human-Like?},
address = {Tokyo, Japan},
year = {2013}
}
Robotics & HRI
Mohammad, Y., & Nishida, T. (2010). Unsupervised Learning of Interactive Behavior for HRI. RSS 2010 Workshop on Learning for Human-Robot Interaction Modeling.
@inproceedings{mohammad2010unsupervisedhri,
keywords = {workshop},
focus = {robotics},
author = {Mohammad, Yasser and Nishida, Toyoaki},
title = {Unsupervised Learning of Interactive Behavior for {HRI}},
booktitle = {RSS 2010 Workshop on Learning for Human-Robot Interaction Modeling},
address = {Zaragoza, Spain},
year = {2010}
}
In this paper, we present our efforts toward building interactive robots that can learn how to interact naturally with human partners in different environments and contexts. The main feature of our approach is that it relies completely on unsupervised learning and time series analysis techniques that allow the robot to build its own interaction protocol representation from the bottom up. The final controller of the robot learned this way is a hierarchy of either dynamical systems or probabilistic networks with complexity that is automatically adjusted to the interaction protocol to be learned. We report two examples of applying this technique to learn an explicit interaction protocol in a master-slave settings (guided navigation) and an implicit protocol in a teammate settings (a listener robot).
Robotics & HRI
Mohammad, Y., Okada, S., & Nishida, T. (2010). Autonomous Development of Gaze Control for Natural Human-Robot Interaction. IUI 2010 Workshop on Eye Gaze in Intelligent Human-Machine Interaction, 63–70.
@inproceedings{mohammad2010gazedevelopment,
keywords = {workshop},
focus = {robotics},
author = {Mohammad, Yasser and Okada, Shogo and Nishida, Toyoaki},
title = {Autonomous Development of Gaze Control for Natural Human-Robot Interaction},
booktitle = {IUI 2010 Workshop on Eye Gaze in Intelligent Human-Machine Interaction},
pages = {63--70},
publisher = {ACM},
year = {2010}
}
Gaze behavior is one of the most important nonverbal behaviors during human-human close encounters. For this reason, many researchers in natural human-robot interaction focus on developing robots that can achieve human-like gaze behavior. Many approaches have been proposed to achieve this natural gaze behavior based on accurate analysis of human behaviors during natural interactions. One limitation of most available approaches is that the behavior is hardwired to the robot and learning techniques are used only, if ever, for adjusting the parameters of the behavior. In this paper we propose and evaluate a different approach in which the robot natural gaze behavior by watching natural interactions between humans. The proposed approach uses the L iEICA architecture developed by the authors and is completely unsupervised which leads to grounded behavior. We compare the resulting gaze controller with a state-of-the-art gaze controller that achieved human-like behavior and show that the proposed approach leads to a more natural gaze behavior based on subjective evaluations of subjects.
Robotics & HRI
Mohammad, Y. (2008). Toward Learning Interactive Behavior. IEEE ICRA 2008 Workshop on New Vistas and Challenges in Telerobotics (NEWHRI).
@inproceedings{mohammad2008towardlearning,
keywords = {workshop},
focus = {robotics},
author = {Mohammad, Yasser},
title = {Toward Learning Interactive Behavior},
booktitle = {IEEE ICRA 2008 Workshop on New Vistas and Challenges in Telerobotics (NEWHRI)},
address = {Pasadena, CA, USA},
year = {2008}
}
This document is divided into two parts. The first part details the authors opinion on the three questions posed by the ICRA NEWHRI 2008 workshop. The second part represents a brief introduction to the research conducted by the author toward realization of autonomous robots that can learn interactive behavior through real world interactions with humans.
Robotics & HRI
Mohammad, Y., & Nishida, T. (2008). Toward Agents that Can Learn Nonverbal Interactive Behavior. IAPR Workshop on Cognitive Information Processing (CIP 2008), 164–169.
@inproceedings{mohammad2008nonverbal,
keywords = {workshop},
focus = {robotics},
author = {Mohammad, Yasser and Nishida, Toyoaki},
title = {Toward Agents that Can Learn Nonverbal Interactive Behavior},
booktitle = {IAPR Workshop on Cognitive Information Processing (CIP 2008)},
pages = {164--169},
address = {Santorini, Greece},
year = {2008}
}
Humans are social agents and the social dimension is an important aspect of human cognition. One challenge facing the realization of artifacts and artificial agents that posses human-like cognition abilities is to implement human-like interactive capabilities into them. Natural Language Processing is one of the earliest applications of AI techniques because of the importance of language in shaping human cognitive and interactive capabilities. Nevertheless nonverbal communication is starting to gain more importance specially in the domains of HRI and ECA because natural human-human communications are known to utilize a variety of nonverbal interaction protocols. This paper proposes a new adaptation algorithm for interactive agents that aims to develop agents that can learn and adapt their theory of mind concerning nonverbal interaction in real-time during actual interactions. The proposed method utilizes elements of the theory of theory and the theory of simulation to guide the adaptation process. A proof of concept simulation experiment with the proposed system is also illustrated.
Robotics & HRI
Mohammad, Y., & Nishida, T. (2008). Natural Listening Robot for AAL Applications. First International Workshop on IUI for Ambient Assisted Living (IUI4AAL 2008).
@inproceedings{mohammad2008aal,
keywords = {workshop},
focus = {robotics},
author = {Mohammad, Yasser and Nishida, Toyoaki},
title = {Natural Listening Robot for {AAL} Applications},
booktitle = {First International Workshop on IUI for Ambient Assisted Living (IUI4AAL 2008)},
address = {Canary Islands, Spain},
year = {2008}
}
The increase in the proportion of senior citizens to the total population is both a challenge and an opportunity for industrial societies. To meet this challenge an increasing research effort is being allocated to the area of Ambient Assisted Living (AAL). Most of the research in this area is concerned with designing novel interfaces and novel services based on embedded intelligence targeting the elderly, and architectures to integrate heterogeneous sensors and solutions. In this paper we argue that a mobile robot companion can provide benefits to this program that are beyond what is possible using only embedded intelligence techniques. We further argue for the benefits of utilizing the robots as knowledge media paradigm for AAL systems. This paper then describes an ongoing effort to realize a humanoid robot that can use human like nonverbal behavior to give the interacting human a listening experiment as a first step for realizing a useful knowledge media companion robot for AAL applications. The details of a proof of applicability study of our approach is given and discussed in relation to AAL.
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2007). A New, HRI Inspired, View of Intention and Intention Communication. AAAI-07 Workshop on Human Implications of Human-Robot Interaction, 21–27.
@inproceedings{mohammad2007hriintention,
keywords = {workshop},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {A New, {HRI} Inspired, View of Intention and Intention Communication},
booktitle = {AAAI-07 Workshop on Human Implications of Human-Robot Interaction},
pages = {21--27},
address = {Vancouver, Canada},
year = {2007}
}
Intention is one of the most important concepts in understanding human subjective existence as well as in creating naturally interacting artifacts and robots. Currently intention is modeled as a fixed yet may be unknown value that affects the behavior of humans and is communicated in natural interactions. This folk psychology inspired vision is underlying the understanding of intention in AI, and HRI. Despite this wide utilization, this view of intention is challenged by recent results in experimental psychology and neuroscience. The contribution of this paper is two fold: 1) A new theory for understanding both intention ( ) and intention communication ( ) is proposed and compared with the traditional view. 2) Three concrete systems based on the proposed view of intention in the areas of HRI, Intelligent User Interfaces, and Robotic Control Architectures are introduced with comparisons to traditional view based systems. This novelle approach to intention can inform research in HRI toward more naturally interactive robots and psychology toward a better understanding of our own intentions.
Robotics & HRI
Mohammad, Y. F. O., & Nishida, T. (2007). EICA: Combining Interactivity with Autonomy for Social Robots. International Workshop on Social Intelligence Design (SID 2007), 227–236.
@inproceedings{mohammad2007eica,
keywords = {workshop},
focus = {robotics},
author = {Mohammad, Yasser F. O. and Nishida, Toyoaki},
title = {{EICA}: Combining Interactivity with Autonomy for Social Robots},
booktitle = {International Workshop on Social Intelligence Design (SID 2007)},
pages = {227--236},
address = {Trento, Italy},
year = {2007}
}
The success of social robots in achieving natural coexistence with humans depends on both their level of autonomy and their interactive abilities. Although many robotic architectures have been suggested and many researchers have focused on human-robot interaction, a robotic architecture that can effectively combine interactivity and autonomy is still unavailable. In this paper a robotic architecture called EICA (Embedded Interactive Control Architecture) is presented that tries to fill this gap. The proposed architecture can help robotic designers in creating sociable robots that can combine both natural interactivity with humans and reactive fast response to the physical and social environmental changes. The resulting robots can then be more socially intelligent than the robots designed using current deliberative, reactive or even hybrid architectures. [Keywords:] Embodiment, EICA, Social Robotics, HRI
Mohammad, Y., & Nishida, T. (2006). Interactive Perception for Amplification of Intended Behavior in Complex Noisy Environment. International Workshop on Social Intelligence Design (SID 2006), 173–187.
@inproceedings{mohammad2006interactiveperception,
keywords = {workshop},
author = {Mohammad, Yasser and Nishida, Toyoaki},
title = {Interactive Perception for Amplification of Intended Behavior in Complex Noisy Environment},
booktitle = {International Workshop on Social Intelligence Design (SID 2006)},
pages = {173--187},
address = {Osaka, Japan},
year = {2006}
}
Local Journals
ML & Time-Series
Ashraf, Z. M., Mohamed, Y. F., & Elsemman, I. E. (2022). Gemminer: Text Mining Tool for Genome-Scale Metabolic Model. Assiut University Journal of Multidisciplinary Scientific Research, 51(3), 358–373.
@article{ashraf2022gemminer,
keywords = {localjournal},
focus = {timeseries},
author = {Ashraf, Zaynab M. and Mohamed, Yasser F. and Elsemman, Ibrahim E.},
title = {Gemminer: Text Mining Tool for Genome-Scale Metabolic Model},
journal = {Assiut University Journal of Multidisciplinary Scientific Research},
volume = {51},
number = {3},
pages = {358--373},
year = {2022}
}
ML & Time-Series
Gad-Elrab, A. A. S., Mohammad, Y. F. O., & El-Melegy, M. T. (2020). Face Recognition from Small Datasets Using Kernel Selection of Gabor Features. Journal of Engineering Sciences, Assiut University, 48(6), 1051–1071.
@article{gadelrab2020face,
keywords = {localjournal},
focus = {timeseries},
author = {Gad-Elrab, Alyaa A. S. and Mohammad, Yasser F. O. and El-Melegy, Moumen T.},
title = {Face Recognition from Small Datasets Using Kernel Selection of Gabor Features},
journal = {Journal of Engineering Sciences, Assiut University},
volume = {48},
number = {6},
pages = {1051--1071},
year = {2020},
month = nov
}
Local Conferences
Others
ML & Time-Series
Heracleous, P., Mohammad, Y., & Yoneyama, A. (2018). Feature Selection Using Deep Neural Networks for Speech Emotion Recognition. Acoustical Society of Japan Autumn Meeting.
@inproceedings{heracleous2018featureselection,
keywords = {other},
focus = {timeseries},
author = {Heracleous, Panikos and Mohammad, Yasser and Yoneyama, Akio},
title = {Feature Selection Using Deep Neural Networks for Speech Emotion Recognition},
booktitle = {Acoustical Society of Japan Autumn Meeting},
address = {Oita, Japan},
year = {2018}
}
Robotics & HRI
Mohammad, Y. (2008). Mutual Intention in Human-Robot Teams. Proceedings of the First Egyptian Japanese International Symposium on Science and Technology (EJISST 2008), 122.
@inproceedings{mohammad2008mutualintention,
keywords = {other},
focus = {robotics},
author = {Mohammad, Yasser},
title = {Mutual Intention in Human-Robot Teams},
booktitle = {Proceedings of the First Egyptian Japanese International Symposium on Science and Technology (EJISST 2008)},
pages = {122},
address = {Tokyo, Japan},
year = {2008}
}