Socially-Conditioned Task Reasoning for a Virtual Tutoring Agent
Saved in:
| Title: | Socially-Conditioned Task Reasoning for a Virtual Tutoring Agent |
|---|---|
| Language: | English |
| Authors: | Zian Zhao, Michael Madaio, Florian Pecune, Yoichi Matsuyama, Justine Cassell |
| Source: | Grantee Submission. 2018. |
| Peer Reviewed: | Y |
| Page Count: | 3 |
| Publication Date: | 2018 |
| Sponsoring Agency: | National Science Foundation (NSF) Institute of Education Sciences (ED) |
| Contract Number: | 1523162 R305B150008 |
| Document Type: | Speeches/Meeting Papers Reports - Research |
| Descriptors: | Tutors, Peer Teaching, Programmed Tutoring, Intelligent Tutoring Systems, Computer Assisted Instruction, Interpersonal Competence, Social Cognition, Thinking Skills, Interpersonal Relationship, Artificial Intelligence |
| Abstract: | Virtual agents have been shown to be more effective when incorporating social factors such as trust into task action selection. However, there has been less work on how virtual tutoring agents can incorporate social factors into pedagogical action selection. We propose and evaluate how a socially-conditioned task reasoner for a virtual pedagogical agent can incorporate both task and social factors into task reasoning. Our work contributes to the autonomous agent community by providing further evidence that incorporating information about dyadic social factors (e.g. rapport) can be beneficial for agents' task reasoning in the case of a tutoring agent. [This paper was published in: "Proc. of the 17th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2018), Stockholm, Sweden, July 10-15, 2018," edited by M. Dastani, G. Sukthankar, E. André, S. Koenig, International Foundation for Autonomous Agents and Multiagent Systems, 2018, pp. 2265-2267. Additional funding provided by the IT R&D program of MSIP/IITP.] |
| Abstractor: | As Provided |
| IES Funded: | Yes |
| Entry Date: | 2024 |
| Accession Number: | ED663182 |
| Database: | ERIC |
| Abstract: | Virtual agents have been shown to be more effective when incorporating social factors such as trust into task action selection. However, there has been less work on how virtual tutoring agents can incorporate social factors into pedagogical action selection. We propose and evaluate how a socially-conditioned task reasoner for a virtual pedagogical agent can incorporate both task and social factors into task reasoning. Our work contributes to the autonomous agent community by providing further evidence that incorporating information about dyadic social factors (e.g. rapport) can be beneficial for agents' task reasoning in the case of a tutoring agent. [This paper was published in: "Proc. of the 17th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2018), Stockholm, Sweden, July 10-15, 2018," edited by M. Dastani, G. Sukthankar, E. André, S. Koenig, International Foundation for Autonomous Agents and Multiagent Systems, 2018, pp. 2265-2267. Additional funding provided by the IT R&D program of MSIP/IITP.] |
|---|
