Pei-hao Su, Pawel Budzianowski, Stefan Ultes, Milica Gasic, Steve J. Young. Sample-efficient Actor-Critic Reinforcement Learning with Supervised Data for Dialogue Management. In Kristiina Jokinen, Manfred Stede, David DeVault, Annie Louis, editors, Proceedings of the 18th Annual SIGdial Meeting on Discourse and Dialogue, Saarbrücken, Germany, August 15-17, 2017. pages 147-157, Association for Computational Linguistics, 2017. [doi]
Abstract is missing.