Dialogue Learning With Human-In-The-Loop
This addresses the challenge of making dialogue systems more adaptive and effective for users, though it appears incremental by building on existing reinforcement learning methods.
The paper tackled the problem of improving conversational agents through online interaction with humans, using reinforcement learning where the bot learns from teacher feedback on its responses. The approach was validated with real experiments on Mechanical Turk.
An important aspect of developing conversational agents is to give a bot the ability to improve through communicating with humans and to learn from the mistakes that it makes. Most research has focused on learning from fixed training sets of labeled data rather than interacting with a dialogue partner in an online fashion. In this paper we explore this direction in a reinforcement learning setting where the bot improves its question-answering ability from feedback a teacher gives following its generated responses. We build a simulator that tests various aspects of such learning in a synthetic environment, and introduce models that work in this regime. Finally, real experiments with Mechanical Turk validate the approach.