A Network-based End-to-End Trainable Task-oriented Dialogue System

Abstract

Teaching machines to accomplish tasks by conversing naturally with humans is challenging. Currently, developing task-oriented dialogue systems requires creating multiple components and typically this involves either a large amount of handcrafting, or acquiring costly labelled datasets to solve a statistical learning problem for each component. In this work we introduce a neural network-based text-in, text-out end-to-end trainable goal-oriented dialogue system along with a new way of collecting dialogue data based on a novel pipe-lined Wizard-of-Oz framework. This approach allows us to develop dialogue systems easily and without making too many assumptions about the task at hand. The results show that the model can converse with human subjects naturally whilst helping them to accomplish tasks in a restaurant search domain.

published at EACL 2017

References in corpus (3)

• Sequence to Sequence Learning with Neural Networks
• Neural Responding Machine for Short-Text Conversation
• On-line Active Reward Learning for Policy Optimisation in Spoken Dialogue Systems

Cited by in corpus (75)

• Wizard of Wikipedia: Knowledge-Powered Conversational agents
• Deep Reinforcement Learning for Dialogue Generation
• The Natural Language Decathlon: Multitask Learning as Question Answering
• A Survey Of Cross-lingual Word Embedding Models
• Efficient Natural Language Response Suggestion for Smart Reply
• Adversarial Learning for Neural Dialogue Generation
• Natural Language Processing Advancements By Deep Learning: A Survey
• A Simple Language Model for Task-Oriented Dialogue
• Rasa: Open Source Language Understanding and Dialogue Management
• Towards End-to-End Learning for Dialog State Tracking and Management using Deep Reinforcement Learning
• LAMOL: LAnguage MOdeling for Lifelong Language Learning
• Two are Better than One: An Ensemble of Retrieval- and Generation-Based Dialog Systems
• Generative Deep Neural Networks for Dialogue: A Short Review
• A Context-aware Natural Language Generator for Dialogue Systems
• Unsupervised Text Style Transfer using Language Models as Discriminators
• CoSQL: A Conversational Text-to-SQL Challenge Towards Cross-Domain Natural Language Interfaces to Databases
• Deep Learning Based Chatbot Models
• Recent advances in conversational NLP : Towards the standardization of Chatbot building
• Towards End-to-End Reinforcement Learning of Dialogue Agents for Information Access
• ConvLab: Multi-Domain End-to-End Dialog System Platform
• Few-shot Natural Language Generation for Task-Oriented Dialog
• Few-Shot Generalization Across Dialogue Tasks
• Towards Explainable and Controllable Open Domain Dialogue Generation with Dialogue Acts
• Training Millions of Personalized Dialogue Agents
• Toward Continual Learning for Conversational Agents
• STAR: A Schema-Guided Dialog Dataset for Transfer Learning
• Personalizing a Dialogue System with Transfer Reinforcement Learning
• Rethinking Action Spaces for Reinforcement Learning in End-to-end Dialog Agents with Latent Variable Models
• Efficient Dialogue State Tracking by Selectively Overwriting Memory
• GuessWhat?! Visual object discovery through multi-modal dialogue
• MALA: Cross-Domain Dialogue Generation with Action Learning
• Zero-Shot Dialog Generation with Cross-Domain Latent Actions
• User Modeling for Task Oriented Dialogues
• Dialog state tracking, a machine reading approach using Memory Network
• End-to-End Trainable Non-Collaborative Dialog System
• Generative Encoder-Decoder Models for Task-Oriented Spoken Dialog Systems with Chatting Capability
• Modeling Text-visual Mutual Dependency for Multi-modal Dialog Generation
• Unsupervised Dialog Structure Learning
• Surprising Negative Results for Generative Adversarial Tree Search
• Show Us the Way: Learning to Manage Dialog from Demonstrations
• Sentiment Adaptive End-to-End Dialog Systems
• Reference-Aware Language Models
• Semantic Refinement GRU-based Neural Language Generation for Spoken Dialogue Systems
• Deep Active Learning for Dialogue Generation
• Deep Reinforcement Learning with a Combinatorial Action Space for Predicting Popular Reddit Threads
• How Managers Perceive AI-Assisted Conversational Training for Workplace Communication
• Reinforcement Learning with External Knowledge and Two-Stage Q-functions for Predicting Popular Reddit Threads
• Generative Visual Dialogue System via Adaptive Reasoning and Weighted Likelihood Estimation
• RADDLE: An Evaluation Benchmark and Analysis Platform for Robust Task-oriented Dialog Systems
• LD-SDS: Towards an Expressive Spoken Dialogue System based on Linked-Data
• Multipurpose Intelligent Process Automation via Conversational Assistant
• Towards Solving Text-based Games by Producing Adaptive Action Spaces
• Cross-domain Dialogue Policy Transfer via Simultaneous Speech-act and Slot Alignment
• Task-oriented Dialogue Systems: performance vs. quality-optima, a review
• Data Augmentation for Copy-Mechanism in Dialogue State Tracking
• DEBACER: a method for slicing moderated debates
• Investigation of Sentiment Controllable Chatbot
• Training like Playing: A Reinforcement Learning And Knowledge Graph-based framework for building Automatic Consultation System in Medical Field
• SYNERGY: Building Task Bots at Scale Using Symbolic Knowledge and Machine Teaching
• Interactive Classification by Asking Informative Questions
• End-to-End Joint Learning of Natural Language Understanding and Dialogue Manager
• Knowledge Grounded Conversational Symptom Detection with Graph Memory Networks
• Achieving Fluency and Coherency in Task-oriented Dialog
• STN4DST: A Scalable Dialogue State Tracking based on Slot Tagging Navigation
• A Short Review on Data Modelling for Vector Fields
• Learning End-to-End Goal-Oriented Dialog with Multiple Answers
• Comprehensive Study: How the Context Information of Different Granularity Affects Dialogue State Tracking?
• Point or Generate Dialogue State Tracker
• AI-Powered Text Generation for Harmonious Human-Machine Interaction: Current State and Future Directions
• Incremental Learning from Scratch for Task-Oriented Dialogue Systems
• Self-Attentional Models Application in Task-Oriented Dialogue Generation Systems
• Generative Dialog Policy for Task-oriented Dialog Systems
• Integrating User and Agent Models: A Deep Task-Oriented Dialogue System
• KddRES: A Multi-level Knowledge-driven Dialogue Dataset for Restaurant Towards Customized Dialogue System
• End to End Dialogue Transformer