Target Entropy Annealing for Discrete Soft Actor-Critic

Soft Actor-Critic (SAC) is considered the state-of-the-art algorithm in continuous action space settings. It uses the maximum entropy framework for efficiency and stability, and applies a heuristic temperature Lagrange term to tune the temperature \(\alpha\), which determines how "soft" th...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2021-12
Main Authors: Xu, Yaosheng, Hu, Dailin, Liang, Litian, McAleer, Stephen, Abbeel, Pieter, Fox, Roy
Format: Article
Language:English
Subjects:
Algorithms
Annealing
Empirical analysis
Entropy
Maximum entropy
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Soft Actor-Critic (SAC) is considered the state-of-the-art algorithm in continuous action space settings. It uses the maximum entropy framework for efficiency and stability, and applies a heuristic temperature Lagrange term to tune the temperature \(\alpha\), which determines how "soft" the policy should be. It is counter-intuitive that empirical evidence shows SAC does not perform well in discrete domains. In this paper we investigate the possible explanations for this phenomenon and propose Target Entropy Scheduled SAC (TES-SAC), an annealing method for the target entropy parameter applied on SAC. Target entropy is a constant in the temperature Lagrange term and represents the target policy entropy in discrete SAC. We compare our method on Atari 2600 games with different constant target entropy SAC, and analyze on how our scheduling affects SAC.
ISSN:2331-8422