Learning End-to-End Goal-Oriented Dialog

Abstract

Traditional dialog systems used in goal-oriented applications require a lot of domain-specific handcrafting, which hinders scaling up to new domains. End-to-end dialog systems, in which all components are trained from the dialogs themselves, escape this limitation. But the encouraging success recently obtained in chit-chat dialog may not carry over to goal-oriented settings. This paper proposes a testbed to break down the strengths and shortcomings of end-to-end dialog systems in goal-oriented applications. Set in the context of restaurant reservation, our tasks require manipulating sentences and symbols, so as to properly conduct conversations, issue API calls and use the outputs of such calls. We show that an end-to-end dialog system based on Memory Networks can reach promising, yet imperfect, performance and learn to perform non-trivial operations. We confirm those results by comparing our system to a hand-crafted slot-filling baseline on data from the second Dialog State Tracking Challenge (Henderson et al., 2014a). We show similar result patterns on data extracted from an online concierge service.

Accepted as a conference paper at ICLR 2017

References in corpus (6)

• Efficient Estimation of Word Representations in Vector Space
• A Neural Conversational Model
• How NOT To Evaluate Your Dialogue System: An Empirical Study of Unsupervised Evaluation Metrics for Dialogue Response Generation
• A Survey of Available Corpora for Building Data-Driven Dialogue Systems
• Learning from Real Users: Rating Dialogue Success with Neural Networks for Reinforcement Learning in Spoken Dialogue Systems
• Reward Shaping with Recurrent Neural Networks for Speeding up On-Line Policy Learning in Spoken Dialogue Systems

Cited by in corpus (45)

• A Survey on Dialogue Systems: Recent Advances and New Frontiers
• Natural Language Processing Advancements By Deep Learning: A Survey
• An End-to-End Trainable Neural Network Model with Belief Tracking for Task-Oriented Dialog
• Best of Both Worlds: Transferring Knowledge from Discriminative Learning to a Generative Visual Dialog Model
• Query-Reduction Networks for Question Answering
• Generative Deep Neural Networks for Dialogue: A Short Review
• Attentive Memory Networks: Efficient Machine Reading for Conversational Search
• Deep Learning Based Chatbot Models
• Towards End-to-End Reinforcement Learning of Dialogue Agents for Information Access
• BiToD: A Bilingual Multi-Domain Dataset For Task-Oriented Dialogue Modeling
• Few-Shot Generalization Across Dialogue Tasks
• Modelling Hierarchical Structure between Dialogue Policy and Natural Language Generator with Option Framework for Task-oriented Dialogue System
• Energy-Based Hindsight Experience Prioritization
• Toward Continual Learning for Conversational Agents
• AI Research Considerations for Human Existential Safety (ARCHES)
• Learning Conversational Systems that Interleave Task and Non-Task Content
• Texar: A Modularized, Versatile, and Extensible Toolkit for Text Generation
• Shaping the Narrative Arc: An Information-Theoretic Approach to Collaborative Dialogue
• Generative Encoder-Decoder Models for Task-Oriented Spoken Dialog Systems with Chatting Capability
• Alquist 2.0: Alexa Prize Socialbot Based on Sub-Dialogue Models
• Sentiment Adaptive End-to-End Dialog Systems
• Gated End-to-End Memory Networks
• Reference-Aware Language Models
• Natural Language Generation with Neural Variational Models
• Natural language understanding for task oriented dialog in the biomedical domain in a low resources context
• Contextual Out-of-Domain Utterance Handling With Counterfeit Data Augmentation
• Prototypical Q Networks for Automatic Conversational Diagnosis and Few-Shot New Disease Adaption
• Conversational Recommender System
• The Rapidly Changing Landscape of Conversational Agents
• End-to-End Knowledge-Routed Relational Dialogue System for Automatic Diagnosis
• Knowledge-incorporating ESIM models for Response Selection in Retrieval-based Dialog Systems
• Recurrent Neural Networks for Dialogue State Tracking
• Finding Dominant User Utterances And System Responses in Conversations
• Combining Textual Content and Structure to Improve Dialog Similarity
• Training like Playing: A Reinforcement Learning And Knowledge Graph-based framework for building Automatic Consultation System in Medical Field
• Hybrid Code Networks using a convolutional neural network as an input layer achieves higher turn accuracy
• Image-based Natural Language Understanding Using 2D Convolutional Neural Networks
• Knowledge-based end-to-end memory networks
• Achieving Fluency and Coherency in Task-oriented Dialog
• Selective Attention Encoders by Syntactic Graph Convolutional Networks for Document Summarization
• Improving Search through A3C Reinforcement Learning based Conversational Agent
• Building a Legal Dialogue System: Development Process, Challenges and Opportunities
• Learning Syntactic and Dynamic Selective Encoding for Document Summarization
• Integrating User and Agent Models: A Deep Task-Oriented Dialogue System
• AI-Powered Text Generation for Harmonious Human-Machine Interaction: Current State and Future Directions