Meta Dialogue Policy Learning

2020/06/03 by Yumo Xu, Xu, Yumo, Chenguang Zhu +5 · 1 citation
Computer Science · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Multimodal Machine Learning Applications #Speech and dialogue systems #Topic Modeling

paper · pdf · doi:10.48550/arxiv.2006.02588

openalex publication_date 2020/06/03 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Dialog policy determines the next-step actions for agents and hence is central to a dialogue system. However, when migrated to novel domains with little data, a policy model can fail to adapt due to insufficient interactions with the new environment. We propose Deep Transferable Q-Network (DTQN) to utilize shareable low-level signals between domains, such as dialogue acts and slots. We decompose the state and action representation space into feature subspaces corresponding to these low-level components to facilitate cross-domain knowledge transfer. Furthermore, we embed DTQN in a meta-learning framework and introduce Meta-DTQN with a dual-replay mechanism to enable effective off-policy training and adaptation. In experiments, our model outperforms baseline models in terms of both success rate and dialogue efficiency on the multi-domain dialogue dataset MultiWOZ 2.0.

Citations

• Playing Atari with Deep Reinforcement Learning
• Strategic Dialogue Management via Deep Reinforcement Learning
• Learning to reinforcement learn
• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Recasting Gradient-Based Meta-Learning as Hierarchical Bayes
• Efficient Off-Policy Meta-Reinforcement Learning via Probabilistic\n Context Variables
• Domain Adaptive Dialog Generation via Meta Learning
• BBQ-Networks: Efficient Exploration in Deep Reinforcement Learning for Task-Oriented Dialogue Systems

Cited by

• Learning to Learn End-to-End Goal-Oriented Dialog From Related Dialog\n Tasks

Related

• Deep Reinforcement Learning for Multi-Domain Dialogue Systems
• Domain Adaptation in Dialogue Systems using Transfer and Meta-Learning
• Learning Dialog Policies from Weak Demonstrations
• Towards Learning Transferable Conversational Skills using Multi-dimensional Dialogue Modelling
• Cross-domain Dialogue Policy Transfer via Simultaneous Speech-act and Slot Alignment