Towards End-to-End Reinforcement Learning of Dialogue Agents for Information Access
This work addresses the challenge of end-to-end training for goal-oriented dialogue agents in information access, offering a solution that improves performance but is incremental in nature.
The paper tackles the problem of training neural dialogue agents for knowledge base access without symbolic queries, which break differentiability, by introducing a soft posterior distribution over entities and integrating it with reinforcement learning, achieving higher task success rates and rewards in simulations and with real users.
This paper proposes KB-InfoBot -- a multi-turn dialogue agent which helps users search Knowledge Bases (KBs) without composing complicated queries. Such goal-oriented dialogue agents typically need to interact with an external database to access real-world knowledge. Previous systems achieved this by issuing a symbolic query to the KB to retrieve entries based on their attributes. However, such symbolic operations break the differentiability of the system and prevent end-to-end training of neural dialogue agents. In this paper, we address this limitation by replacing symbolic queries with an induced "soft" posterior distribution over the KB that indicates which entities the user is interested in. Integrating the soft retrieval process with a reinforcement learner leads to higher task success rate and reward in both simulations and against real users. We also present a fully neural end-to-end agent, trained entirely from user feedback, and discuss its application towards personalized dialogue agents. The source code is available at https://github.com/MiuLab/KB-InfoBot.