Two Approaches to Building Collaborative, Task-Oriented Dialog Agents through Self-Play
Not on the map. Why?
Abstract (arXiv)
Task-oriented dialog systems are often trained on human/human dialogs, such as collected from Wizard-of-Oz interfaces. However, human/human corpora are frequently too small for supervised training to be effective. This paper investigates two approaches to training agent-bots and user-bots through self-play, in which they autonomously explore an API environment, discovering communication strategies that enable them to solve the task. We give empirical results for both reinforcement learning and game-theoretic equilibrium finding.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 12 of its 27 references that are in the map. First posted on arXiv 20 Sep 2021.
Builds on
12 of 27 references.
References · 12
- 2017Deal or No Deal? End-to-End Learning for Negotiation Dialoguescited by 202
- 2019The Hanabi Challenge: A New Frontier for AI Researchcited by 194
- 2018Building a Conversational Agent Overnight with Dialogue Self-Playcited by 70
- 2017RLlib: Abstractions for Distributed Reinforcement Learningcited by 53
- 2019How to Build User Simulators to Train RL-based Dialog Systemscited by 44
- 2019Recommendation as a Communication Game: Self-Supervised Bot-Play for Goal-oriented Dialoguecited by 44
- 2018Neural User Simulation for Corpus-based Policy Optimisation for Spoken Dialogue Systemscited by 42
- 2016A Sequence-to-Sequence Model for User Simulation in Spoken Dialogue Systemscited by 39
- 2018User Modeling for Task Oriented Dialoguescited by 36
- 2018Adversarial Learning of Task-Oriented Neural Dialog Modelscited by 17
- 2019Generating Challenge Datasets for Task-Oriented Conversational Agents through Self-Playcited by 2
- 2020MEEP: An Open-Source Platform for Human-Human Dialog Collection and End-to-End Agent Trainingcited by 2
Built on by
Not cited yet.