Joint System-Wise Optimization for Pipeline Goal-Oriented Dialog System

Abstract

Recent work (Takanobu et al., 2020) proposed the system-wise evaluation on dialog systems and found that improvement on individual components (e.g., NLU, policy) in prior work may not necessarily bring benefit to pipeline systems in system-wise evaluation. To improve the system-wise performance, in this paper, we propose new joint system-wise optimization techniques for the pipeline dialog system. First, we propose a new data augmentation approach which automates the labeling process for NLU training. Second, we propose a novel stochastic policy parameterization with Poisson distribution that enables better exploration and offers a principled way to compute policy gradient. Third, we propose a reward bonus to help policy explore successful dialogs. Our approaches outperform the competitive pipeline systems from Takanobu et al. (2020) by big margins of 12% success rate in automatic system-wise evaluation and of 16% success rate in human evaluation on the standard multi-domain benchmark dataset MultiWOZ 2.1, and also outperform the recent state-of-the-art end-to-end trained model from DSTC9.

arXiv comment13 pages

References in corpus (15)

• Dialogue Act Modeling for Automatic Tagging and Recognition of Conversational Speech
• EDA: Easy Data Augmentation Techniques for Boosting Performance on Text Classification Tasks
• A Simple Language Model for Task-Oriented Dialogue
• Towards End-to-End Learning for Dialog State Tracking and Management using Deep Reinforcement Learning
• SOLOIST: Building Task Bots at Scale with Transfer Learning and Machine Teaching
• The Eighth Dialog System Technology Challenge
• Composite Task-Completion Dialogue Policy Learning via Hierarchical Deep Reinforcement Learning
• SUMBT: Slot-Utterance Matching for Universal and Scalable Belief Tracking
• Rethinking Action Spaces for Reinforcement Learning in End-to-end Dialog Agents with Latent Variable Models
• Dialogue Learning with Human Teaching and Feedback in End-to-End Trainable Task-Oriented Dialogue Systems
• ConvLab-2: An Open-Source Toolkit for Building, Evaluating, and Diagnosing Dialogue Systems
• AuGPT: Auxiliary Tasks and Data Augmentation for End-To-End Dialogue with Pre-Trained Language Models
• Is Your Goal-Oriented Dialog Model Performing Really Well? Empirical Analysis of System-wise Evaluation
• A Hybrid Task-Oriented Dialog System with Domain and Task Adaptive Pretraining
• Robustness Testing of Language Understanding in Task-Oriented Dialog