Adapting a teachable robot's dialog responses using reinforcement learning in teaching conversation

Rachel Love, Edith Law, Philip R. Cohen, Dana Kulic

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

1 Citation (Scopus)

Abstract

Teachable robots can offer benefits to students through the use of social behaviours, such as speech, gaze, and gestures, to promote engagement and learning. Adapting these behaviours can deliver personalised interactions to better suit each individual. There is a growing body of research utilising reinforcement learning in social robotics, however there is limited research in the use of adaptive dialog behaviours for social robots. We propose an adaptive response-selection algorithm for a teachable robot which aims to improve user engagement in the teaching task. The proposed approach uses Q-learning to learn an individualised policy. The algorithm is rewarded according to the time taken per teaching input, and the amount paraphrasing in the user's response. A user study has been conducted to evaluate the algorithm, compared to a method of random response-selection. The results indicate that an adaptive approach learns to select more rewarding actions over time, and personalise to the individual user.

Original languageEnglish
Title of host publication2023 32nd IEEE International Conference on Robot and Human Interactive Communication, RO-MAN 2023
EditorsMinsu Jang, Chung Hyuk Park
Place of PublicationPiscataway NJ USA
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages2541-2548
Number of pages8
ISBN (Electronic)9798350336702
ISBN (Print)9798350336719
DOIs
Publication statusPublished - 2023
EventIEEE International Conference on Robot and Human Interactive Communication 2023 - Busan, Korea, South
Duration: 28 Aug 202331 Aug 2023
Conference number: 32nd
https://ro-man2023.org/main (Website)
https://ieeexplore.ieee.org/xpl/conhome/10309296/proceeding?isnumber=10309265&sortType=vol-only-seq&rowsPerPage=100&pageNumber=1 (Proceedings)

Publication series

NameIEEE International Workshop on Robot and Human Communication, RO-MAN
PublisherIEEE, Institute of Electrical and Electronics Engineers
ISSN (Print)1944-9445
ISSN (Electronic)1944-9437

Conference

ConferenceIEEE International Conference on Robot and Human Interactive Communication 2023
Abbreviated titleRO-MAN 2023
Country/TerritoryKorea, South
CityBusan
Period28/08/2331/08/23
Internet address

Cite this