AI Chat Paper
Note: Please note that the following content is generated by AMiner AI. SciOpen does not take any responsibility related to this content.
{{lang === 'zh_CN' ? '文章概述' : 'Summary'}}
{{lang === 'en_US' ? '中' : 'Eng'}}
Chat more with AI
Article Link
Collect
Show Outline
Outline
Show full outline
Hide outline
Outline
Show full outline
Hide outline
Full Length Article | Open Access

Controlling underestimation bias in reinforcement learning via minmax operation

Fanghui HUANGa,b, Yixin HEb, Yu ZHANGa, Xinyang DENGa, Wen JIANGa( )
School of Electronics and Information, Northwestern Polytechnical University, Xi’an 710129, China
College of Information Science and Engineering, Jiaxing University, Jiaxing 314001, China

☆ Peer review under responsibility of Editorial Committee of CJA.

Show Author Information

Abstract

Obtaining the accurate value estimation and reducing the estimation bias are the key issues in reinforcement learning. However, current methods that address the overestimation problem tend to introduce underestimation, which face a challenge of precise decision-making in many fields. To address this issue, we conduct a theoretical analysis of the underestimation bias and propose the minmax operation, which allow for flexible control of the estimation bias. Specifically, we select the maximum value of each action from multiple parallel state-action networks to create a new state-action value sequence. Then, a minimum value is selected to obtain more accurate value estimations. Moreover, based on the minmax operation, we propose two novel algorithms by combining Deep Q-Network (DQN) and Double DQN (DDQN), named minmax-DQN and minmax-DDQN. Meanwhile, we conduct theoretical analyses of the estimation bias and variance caused by our proposed minmax operation, which show that this operation significantly improves both underestimation and overestimation biases and leads to the unbiased estimation. Furthermore, the variance is also reduced, which is helpful to improve the network training stability. Finally, we conduct numerous comparative experiments in various environments, which empirically demonstrate the superiority of our method.

References

【1】
【1】
 
 
Chinese Journal of Aeronautics
Pages 406-417

{{item.num}}

Comments on this article

Go to comment

< Back to all reports

Review Status: {{reviewData.commendedNum}} Commended , {{reviewData.revisionRequiredNum}} Revision Required , {{reviewData.notCommendedNum}} Not Commended Under Peer Review

Review Comment

Close
Close
Cite this article:
HUANG F, HE Y, ZHANG Y, et al. Controlling underestimation bias in reinforcement learning via minmax operation. Chinese Journal of Aeronautics, 2024, 37(7): 406-417. https://doi.org/10.1016/j.cja.2024.03.008

962

Views

9

Crossref

8

Web of Science

10

Scopus

0

CSCD

Received: 20 July 2023
Revised: 25 September 2023
Accepted: 06 November 2023
Published: 11 March 2024
© 2024 Chinese Society of Aeronautics and Astronautics

This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).