Mutual Modelling in Human-Robot Interaction
Event details
| Date | 26.06.2019 |
| Hour | 12:00 › 14:00 |
| Speaker | Utku Norman |
| Location | |
| Category | Conferences - Seminars |
EDIC candidacy exam
Exam president: Prof. Aude Billard
Thesis advisor: Prof. Pierre Dillenbourg
Co-examiner: Prof. Frédéric Kaplan
Abstract
In the context of collaborative learning with a robot,
the ability to build and maintain a mental model of the other
is essential to comprehend the other and react properly. As
humans, this skill of mutual modeling is performed during most
interactions by attributing goals and beliefs to others. Thus, we
consider how to equip a robot with mutual modelling capabilities
in order to increase the quality of the dialogue and hopefully
the learning gains. As background, we present three papers
on i) methodologies for an agent to model another agent, ii) a
Bayesian framework for accounting how humans model an agent,
and iii) a study on grounding, i.e. the process of constructing a
mutual understanding, in collaborative problem solving.
Background papers
Autonomous agents modelling other agents: A comprehensive survey and open problems. Artificial Intelligence, 258 (May 2018), 66–95. by Albrecht, S. V., & Stone, P.
Rational quantitative attribution of beliefs, desires and percepts in human mentalizing, by Baker, C. L., Jara-Ettinger, J., Saxe, R., & Tenenbaum, J. B. Nature Human Behaviour, 1(4), 1–10.
Sharing Solutions: Persistence and Grounding in Multimodal Collaborative Problem Solving, by Dillenbourg, P., & Traum, D. (2006). Journal of the Learning Sciences, 15(1), 121–151.
Exam president: Prof. Aude Billard
Thesis advisor: Prof. Pierre Dillenbourg
Co-examiner: Prof. Frédéric Kaplan
Abstract
In the context of collaborative learning with a robot,
the ability to build and maintain a mental model of the other
is essential to comprehend the other and react properly. As
humans, this skill of mutual modeling is performed during most
interactions by attributing goals and beliefs to others. Thus, we
consider how to equip a robot with mutual modelling capabilities
in order to increase the quality of the dialogue and hopefully
the learning gains. As background, we present three papers
on i) methodologies for an agent to model another agent, ii) a
Bayesian framework for accounting how humans model an agent,
and iii) a study on grounding, i.e. the process of constructing a
mutual understanding, in collaborative problem solving.
Background papers
Autonomous agents modelling other agents: A comprehensive survey and open problems. Artificial Intelligence, 258 (May 2018), 66–95. by Albrecht, S. V., & Stone, P.
Rational quantitative attribution of beliefs, desires and percepts in human mentalizing, by Baker, C. L., Jara-Ettinger, J., Saxe, R., & Tenenbaum, J. B. Nature Human Behaviour, 1(4), 1–10.
Sharing Solutions: Persistence and Grounding in Multimodal Collaborative Problem Solving, by Dillenbourg, P., & Traum, D. (2006). Journal of the Learning Sciences, 15(1), 121–151.
Practical information
- General public
- Free