Skill-Critic: Refining Learned Skills for Hierarchical Reinforcement Learning

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Hao, Ce, Weaver, Catherine, Tang, Chen, Kawamoto, Kenta, Tomizuka, Masayoshi, Zhan, Wei
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917719854546944
author Hao, Ce
Weaver, Catherine
Tang, Chen
Kawamoto, Kenta
Tomizuka, Masayoshi
Zhan, Wei
author_facet Hao, Ce
Weaver, Catherine
Tang, Chen
Kawamoto, Kenta
Tomizuka, Masayoshi
Zhan, Wei
contents Hierarchical reinforcement learning (RL) can accelerate long-horizon decision-making by temporally abstracting a policy into multiple levels. Promising results in sparse reward environments have been seen with skills, i.e. sequences of primitive actions. Typically, a skill latent space and policy are discovered from offline data. However, the resulting low-level policy can be unreliable due to low-coverage demonstrations or distribution shifts. As a solution, we propose the Skill-Critic algorithm to fine-tune the low-level policy in conjunction with high-level skill selection. Our Skill-Critic algorithm optimizes both the low-level and high-level policies; these policies are initialized and regularized by the latent space learned from offline demonstrations to guide the parallel policy optimization. We validate Skill-Critic in multiple sparse-reward RL environments, including a new sparse-reward autonomous racing task in Gran Turismo Sport. The experiments show that Skill-Critic's low-level policy fine-tuning and demonstration-guided regularization are essential for good performance. Code and videos are available at our website: https://sites.google.com/view/skill-critic.
format Preprint
id arxiv_https___arxiv_org_abs_2306_08388
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Skill-Critic: Refining Learned Skills for Hierarchical Reinforcement Learning
Hao, Ce
Weaver, Catherine
Tang, Chen
Kawamoto, Kenta
Tomizuka, Masayoshi
Zhan, Wei
Machine Learning
Artificial Intelligence
Hierarchical reinforcement learning (RL) can accelerate long-horizon decision-making by temporally abstracting a policy into multiple levels. Promising results in sparse reward environments have been seen with skills, i.e. sequences of primitive actions. Typically, a skill latent space and policy are discovered from offline data. However, the resulting low-level policy can be unreliable due to low-coverage demonstrations or distribution shifts. As a solution, we propose the Skill-Critic algorithm to fine-tune the low-level policy in conjunction with high-level skill selection. Our Skill-Critic algorithm optimizes both the low-level and high-level policies; these policies are initialized and regularized by the latent space learned from offline demonstrations to guide the parallel policy optimization. We validate Skill-Critic in multiple sparse-reward RL environments, including a new sparse-reward autonomous racing task in Gran Turismo Sport. The experiments show that Skill-Critic's low-level policy fine-tuning and demonstration-guided regularization are essential for good performance. Code and videos are available at our website: https://sites.google.com/view/skill-critic.
title Skill-Critic: Refining Learned Skills for Hierarchical Reinforcement Learning
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2306.08388