PAC-Bayesian Soft Actor-Critic Learning

Bahareh Tasdighi*, Abdullah Akgül, Manuel Haussmann, Kenny Kazimirzak Brink, Melih Kandemir

*Kontaktforfatter

Publikation: Bidrag til tidsskriftKonferenceartikelForskningpeer review

Abstract

Actor-critic algorithms address the dual goals of reinforcement learning (RL), policy evaluation and improvement via two separate function approximators. The practicality of this approach comes at the expense of training instability, caused mainly by the destructive effect of the approximation errors of the critic on the actor. We tackle this bottleneck by employing an existing Probably Approximately Correct (PAC) Bayesian bound for the first time as the critic training objective of the Soft Actor-Critic (SAC) algorithm. We further demonstrate that online learning performance improves significantly when a stochastic actor explores multiple futures by critic-guided random search. We observe our resulting algorithm to compare favorably against the state-of-the-art SAC implementation on multiple classical control and locomotion tasks in terms of both sample efficiency and regret.

OriginalsprogEngelsk
BogserieProceedings of Machine Learning Research
Vol/bind253
Sider (fra-til)127-145
ISSN2640-3498
StatusUdgivet - 2024
Begivenhed6th Symposium on Advances in Approximate Bayesian Inference, AABI 2024 - Vienna, Østrig
Varighed: 21. jul. 2024 → …

Konference

Konference6th Symposium on Advances in Approximate Bayesian Inference, AABI 2024
Land/OmrådeØstrig
ByVienna
Periode21/07/2024 → …

Fingeraftryk

Dyk ned i forskningsemnerne om 'PAC-Bayesian Soft Actor-Critic Learning'. Sammen danner de et unikt fingeraftryk.

Citationsformater