Task-based dialogue policy learning based on decision transformer

Zhenyou Zhou; Zhibin Liu; Yuhan Liu

doi:10.1117/12.2673682

22 May 2023 Task-based dialogue policy learning based on decision transformer

Zhenyou Zhou, Zhibin Liu, Yuhan Liu

Proceedings Volume 12640, International Conference on Internet of Things and Machine Learning (IoTML 2022); 126401O (2023) https://doi.org/10.1117/12.2673682
Event: International Conference on Internet of Things and Machine Learning (IoTML 2022), 2022, Harbin, China

Abstract

Due to interactions with real users, online reinforcement learning training projects for dialogue agents are expensive. User simulator is an alternative method that is commonly used. However, the environment of a user simulator is not identical to that of a real user, and it cannot provide the atypical and more variegated conversational behavior that is a hallmark of human spontaneity. We employ offline reinforcement learning and Transformer to abstract dialogue policy as a framework for sequence modeling problems, modeling the joint distribution of state, action, and reward sequences to generate optimal dialogue actions. An evaluation of the Multiwoz dataset shows that DT successfully improves the efficiency of DRL dialogue agents and improves dialogue robustness.

Citation Download Citation

Zhenyou Zhou, Zhibin Liu, and Yuhan Liu "Task-based dialogue policy learning based on decision transformer", Proc. SPIE 12640, International Conference on Internet of Things and Machine Learning (IoTML 2022), 126401O (22 May 2023); https://doi.org/10.1117/12.2673682

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available