Saved in:
Bibliographic Details
Main Authors: Chen, Xingguo, Gong, Yu, Yang, Shangdong, Wang, Wenhao
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2411.06396
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915012739596288
author Chen, Xingguo
Gong, Yu
Yang, Shangdong
Wang, Wenhao
author_facet Chen, Xingguo
Gong, Yu
Yang, Shangdong
Wang, Wenhao
contents Fast-converging algorithms are a contemporary requirement in reinforcement learning. In the context of linear function approximation, the magnitude of the smallest eigenvalue of the key matrix is a major factor reflecting the convergence speed. Traditional value-based RL algorithms focus on minimizing errors. This paper introduces a variance minimization (VM) approach for value-based RL instead of error minimization. Based on this approach, we proposed two objectives, the Variance of Bellman Error (VBE) and the Variance of Projected Bellman Error (VPBE), and derived the VMTD, VMTDC, and VMETD algorithms. We provided proofs of their convergence and optimal policy invariance of the variance minimization. Experimental studies validate the effectiveness of the proposed algorithms.
format Preprint
id arxiv_https___arxiv_org_abs_2411_06396
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A Variance Minimization Approach to Temporal-Difference Learning
Chen, Xingguo
Gong, Yu
Yang, Shangdong
Wang, Wenhao
Machine Learning
Artificial Intelligence
Fast-converging algorithms are a contemporary requirement in reinforcement learning. In the context of linear function approximation, the magnitude of the smallest eigenvalue of the key matrix is a major factor reflecting the convergence speed. Traditional value-based RL algorithms focus on minimizing errors. This paper introduces a variance minimization (VM) approach for value-based RL instead of error minimization. Based on this approach, we proposed two objectives, the Variance of Bellman Error (VBE) and the Variance of Projected Bellman Error (VPBE), and derived the VMTD, VMTDC, and VMETD algorithms. We provided proofs of their convergence and optimal policy invariance of the variance minimization. Experimental studies validate the effectiveness of the proposed algorithms.
title A Variance Minimization Approach to Temporal-Difference Learning
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2411.06396