Budgeted Policy Learning for Task-Oriented Dialogue Systems

Zhirui Zhang, Xiujun Li, Jianfeng Gao, Enhong Chen


Abstract
This paper presents a new approach that extends Deep Dyna-Q (DDQ) by incorporating a Budget-Conscious Scheduling (BCS) to best utilize a fixed, small amount of user interactions (budget) for learning task-oriented dialogue agents. BCS consists of (1) a Poisson-based global scheduler to allocate budget over different stages of training; (2) a controller to decide at each training step whether the agent is trained using real or simulated experiences; (3) a user goal sampling module to generate the experiences that are most effective for policy learning. Experiments on a movie-ticket booking task with simulated and real users show that our approach leads to significant improvements in success rate over the state-of-the-art baselines given the fixed budget.
Anthology ID:
P19-1364
Volume:
Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics
Month:
July
Year:
2019
Address:
Florence, Italy
Editors:
Anna Korhonen, David Traum, Lluís Màrquez
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
3742–3751
Language:
URL:
https://aclanthology.org/P19-1364
DOI:
10.18653/v1/P19-1364
Bibkey:
Cite (ACL):
Zhirui Zhang, Xiujun Li, Jianfeng Gao, and Enhong Chen. 2019. Budgeted Policy Learning for Task-Oriented Dialogue Systems. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 3742–3751, Florence, Italy. Association for Computational Linguistics.
Cite (Informal):
Budgeted Policy Learning for Task-Oriented Dialogue Systems (Zhang et al., ACL 2019)
Copy Citation:
PDF:
https://aclanthology.org/P19-1364.pdf