User-Aware Dialogue Management Policies over Attributed Bi-Automata

Abstract

Designing dialogue policies that take user behavior into account is complicated due to user vari- ability and behavioral uncertainty. Attributed Prob- abilistic Finite State Bi-Automata (A-PFSBA) have proven to be a promising framework to develop dia- logue managers that capture the users’ actions in its structure and adapt to them online, yet developing poli- cies robust to high user uncertainty is still challenging. In this paper, the theoretical A-PFSBA dialogue man- agement framework is augmented by formally defining the notation of exploitation policies over its structure. Under such definition, multiple path based policies are implemented, those that take into account external in- formation and those which do not. These policies are evaluated on the Let’s Go corpus, before and after an online learning process whose goal is to update the ini- tial model through the interaction with end-users. In these experiments the impact of user uncertainty and the model structural learning is thoroughly analyzed