A Theoretical Justification for Asymmetric Actor-Critic Algorithms

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lambrechts, Gaspard, Ernst, Damien, Mahajan, Aditya
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915483599503360
author Lambrechts, Gaspard
Ernst, Damien
Mahajan, Aditya
author_facet Lambrechts, Gaspard
Ernst, Damien
Mahajan, Aditya
contents In reinforcement learning for partially observable environments, many successful algorithms have been developed within the asymmetric learning paradigm. This paradigm leverages additional state information available at training time for faster learning. Although the proposed learning objectives are usually theoretically sound, these methods still lack a precise theoretical justification for their potential benefits. We propose such a justification for asymmetric actor-critic algorithms with linear function approximators by adapting a finite-time convergence analysis to this setting. The resulting finite-time bound reveals that the asymmetric critic eliminates error terms arising from aliasing in the agent state.
format Preprint
id arxiv_https___arxiv_org_abs_2501_19116
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Theoretical Justification for Asymmetric Actor-Critic Algorithms
Lambrechts, Gaspard
Ernst, Damien
Mahajan, Aditya
Machine Learning
In reinforcement learning for partially observable environments, many successful algorithms have been developed within the asymmetric learning paradigm. This paradigm leverages additional state information available at training time for faster learning. Although the proposed learning objectives are usually theoretically sound, these methods still lack a precise theoretical justification for their potential benefits. We propose such a justification for asymmetric actor-critic algorithms with linear function approximators by adapting a finite-time convergence analysis to this setting. The resulting finite-time bound reveals that the asymmetric critic eliminates error terms arising from aliasing in the agent state.
title A Theoretical Justification for Asymmetric Actor-Critic Algorithms
topic Machine Learning
url https://arxiv.org/abs/2501.19116