Reinforcement Learning for Realistic Statistical Spoken Dialogue Systems - Beyond Slot-Filling Applications
To date, statistical approaches to dialogue management based on Reinforcement Learning have focused on slot-filling spoken dialogue systems, largely ignoring other types of dialogue. The goal of this project is to extend this line of research to allow automated dialogue policy creation for multiple types of dialogue, including question-answering and negotiation.
since 2011
National Science Foundation
Publications
-
Reinforcement learning of two-issue negotiation dialogue policies. In: 14th Annual SIGdial Meeting on Discourse and Dialogue. Metz, France; In Press.
-
Culture-specific models of negotiation for virtual characters: multi-attribute decision-making based on culture-specific values. AI & Society, Special Issue on Culturally Motivated Virtual Characters. In Press.
-
Full TextReinforcement learning of question-answering dialogue policies for virtual museum guides. In: 13th Annual SIGdial meeting on Discourse and Dialogue. Seoul, South Korea; 2012.
-
Full TextA cultural decision-making model for negotiation based on inverse reinforcement learning. In: 34th Annual Meeting of the Cognitive Science Society. Sapporo, Japan; 2012.