Deep Reinforcement Learning for On-line Dialogue State Tracking

2020/09/22 by Chen, Zhi, Chen, Lu, Zhou, Xiang +1
#Computation and Language (cs.CL) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.2009.10321

Abstract

Dialogue state tracking (DST) is a crucial module in dialogue management. It is usually cast as a supervised training problem, which is not convenient for on-line optimization. In this paper, a novel companion teaching based deep reinforcement learning (DRL) framework for on-line DST optimization is proposed. To the best of our knowledge, this is the first effort to optimize the DST module within DRL framework for on-line task-oriented spoken dialogue systems. In addition, dialogue policy can be further jointly updated. Experiments show that on-line DST optimization can effectively improve the dialogue manager performance while keeping the flexibility of using predefined policy. Joint training of both DST and policy can further improve the performance.

Related

• End-to-End Optimization of Task-Oriented Dialogue Model with Deep Reinforcement Learning
• Towards End-to-End Learning for Dialog State Tracking and Management using Deep Reinforcement Learning
• Domain Transfer in Dialogue Systems without Turn-Level Supervision
• DORA: Toward Policy Optimization for Task-oriented Dialogue System with Efficient Context
• Improving Multi-Domain Task-Oriented Dialogue System with Offline Reinforcement Learning