Actor-Double-Critic: Incorporating Model-Based Critic for Task-Oriented Dialogue Systems

Yen-chen Wu, Bo-Hsiang Tseng, Milica Gasic


Abstract
In order to improve the sample-efficiency of deep reinforcement learning (DRL), we implemented imagination augmented agent (I2A) in spoken dialogue systems (SDS). Although I2A achieves a higher success rate than baselines by augmenting predicted future into a policy network, its complicated architecture introduces unwanted instability. In this work, we propose actor-double-critic (ADC) to improve the stability and overall performance of I2A. ADC simplifies the architecture of I2A to reduce excessive parameters and hyper-parameters. More importantly, a separate model-based critic shares parameters between actions and makes back-propagation explicit. In our experiments on Cambridge Restaurant Booking task, ADC enhances success rates considerably and shows robustness to imperfect environment models. In addition, ADC exhibits the stability and sample-efficiency as significantly reducing the baseline standard deviation of success rates and reaching the 80% success rate with half training data.
Anthology ID:
2020.findings-emnlp.75
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2020
Month:
November
Year:
2020
Address:
Online
Editors:
Trevor Cohn, Yulan He, Yang Liu
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
854–863
Language:
URL:
https://aclanthology.org/2020.findings-emnlp.75
DOI:
10.18653/v1/2020.findings-emnlp.75
Bibkey:
Cite (ACL):
Yen-chen Wu, Bo-Hsiang Tseng, and Milica Gasic. 2020. Actor-Double-Critic: Incorporating Model-Based Critic for Task-Oriented Dialogue Systems. In Findings of the Association for Computational Linguistics: EMNLP 2020, pages 854–863, Online. Association for Computational Linguistics.
Cite (Informal):
Actor-Double-Critic: Incorporating Model-Based Critic for Task-Oriented Dialogue Systems (Wu et al., Findings 2020)
Copy Citation:
PDF:
https://preview.aclanthology.org/naacl24-info/2020.findings-emnlp.75.pdf