Dialog State Tracking with Reinforced Data Augmentation

TL;DR

This paper introduces a reinforcement learning-based data augmentation framework for neural dialog state tracking, significantly enhancing performance with limited training data by generating high-quality, context-aware training instances.

Contribution

It proposes a novel RL-based contextual bandit generator that learns fine-grained augmentation policies and iteratively refines data quality for dialog state tracking.

Findings

Significant performance improvements on WoZ and MultiWoZ datasets.

Effective data augmentation with limited labeled data.

Enhanced dialog state tracking accuracy.

Abstract

Neural dialog state trackers are generally limited due to the lack of quantity and diversity of annotated training data. In this paper, we address this difficulty by proposing a reinforcement learning (RL) based framework for data augmentation that can generate high-quality data to improve the neural state tracker. Specifically, we introduce a novel contextual bandit generator to learn fine-grained augmentation policies that can generate new effective instances by choosing suitable replacements for the specific context. Moreover, by alternately learning between the generator and the state tracker, we can keep refining the generative policies to generate more high-quality training data for neural state tracker. Experimental results on the WoZ and MultiWoZ (restaurant) datasets demonstrate that the proposed framework significantly improves the performance over the state-of-the-art models, especially with limited training data.