Learning Robust Dialog Policies in Noisy Environments

Abstract

Modern virtual personal assistants provide a convenient interface for completing daily tasks via voice commands. An important consideration for these assistants is the ability to recover from automatic speech recognition (ASR) and natural language understanding (NLU) errors. In this paper, we focus on learning robust dialog policies to recover from these errors. To this end, we develop a user simulator which interacts with the assistant through voice commands in realistic scenarios with noisy audio, and use it to learn dialog policies through deep reinforcement learning. We show that dialogs generated by our simulator are indistinguishable from human generated dialogs, as determined by human evaluators. Furthermore, preliminary experimental results show that the learned policies in noisy environments achieve the same execution success rate with fewer dialog turns compared to fixed rule-based policies.

1st Workshop on Conversational AI at NIPS 2017

References in corpus (9)

• Playing Atari with Deep Reinforcement Learning
• Dueling Network Architectures for Deep Reinforcement Learning
• On the Properties of Neural Machine Translation: Encoder-Decoder Approaches
• Deep Reinforcement Learning for Dialogue Generation
• Optimizing Dialogue Management with Reinforcement Learning: Experiments with the NJFun System
• A User Simulator for Task-Completion Dialogues
• Towards End-to-End Learning for Dialog State Tracking and Management using Deep Reinforcement Learning
• Just ASK: Building an Architecture for Extensible Self-Service Spoken Language Understanding
• SimpleDS: A Simple Deep Reinforcement Learning Dialogue System

Cited by in corpus (9)

• Open-Domain Conversational Agents: Current Progress, Open Problems, and Future Directions
• Investigation of Error Simulation Techniques for Learning Dialog Policies for Conversational Error Recovery
• Style Attuned Pre-training and Parameter Efficient Fine-tuning for Spoken Language Understanding
• Towards Personalized Dialog Policies for Conversational Skill Discovery
• Document-editing Assistants and Model-based Reinforcement Learning as a Path to Conversational AI
• Teaching Models new APIs: Domain-Agnostic Simulators for Task Oriented Dialogue
• Alexa Conversations: An Extensible Data-driven Approach for Building Task-oriented Dialogue Systems
• Knowledge Grounded Conversational Symptom Detection with Graph Memory Networks
• Meta learning to classify intent and slot labels with noisy few shot examples