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.
@article{arxiv.2109.09597,
title = {Two Approaches to Building Collaborative, Task-Oriented Dialog Agents through Self-Play},
author = {Arkady Arkhangorodsky and Scot Fang and Victoria Knight and Ajay Nagesh and Maria Ryskina and Kevin Knight},
journal= {arXiv preprint arXiv:2109.09597},
year = {2021}
}