EXTRACT: Efficient Policy Learning by Extracting Transferable Robot Skills from Offline Data

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Zhang, Jesse, Heo, Minho, Liu, Zuxin, Biyik, Erdem, Lim, Joseph J, Liu, Yao, Fakoor, Rasool
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916400066461696
author Zhang, Jesse
Heo, Minho
Liu, Zuxin
Biyik, Erdem
Lim, Joseph J
Liu, Yao
Fakoor, Rasool
author_facet Zhang, Jesse
Heo, Minho
Liu, Zuxin
Biyik, Erdem
Lim, Joseph J
Liu, Yao
Fakoor, Rasool
contents Most reinforcement learning (RL) methods focus on learning optimal policies over low-level action spaces. While these methods can perform well in their training environments, they lack the flexibility to transfer to new tasks. Instead, RL agents that can act over useful, temporally extended skills rather than low-level actions can learn new tasks more easily. Prior work in skill-based RL either requires expert supervision to define useful skills, which is hard to scale, or learns a skill-space from offline data with heuristics that limit the adaptability of the skills, making them difficult to transfer during downstream RL. Our approach, EXTRACT, instead utilizes pre-trained vision language models to extract a discrete set of semantically meaningful skills from offline data, each of which is parameterized by continuous arguments, without human supervision. This skill parameterization allows robots to learn new tasks by only needing to learn when to select a specific skill and how to modify its arguments for the specific task. We demonstrate through experiments in sparse-reward, image-based, robot manipulation environments that EXTRACT can more quickly learn new tasks than prior works, with major gains in sample efficiency and performance over prior skill-based RL. Website at https://www.jessezhang.net/projects/extract/.
format Preprint
id arxiv_https___arxiv_org_abs_2406_17768
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle EXTRACT: Efficient Policy Learning by Extracting Transferable Robot Skills from Offline Data
Zhang, Jesse
Heo, Minho
Liu, Zuxin
Biyik, Erdem
Lim, Joseph J
Liu, Yao
Fakoor, Rasool
Robotics
Artificial Intelligence
Machine Learning
Most reinforcement learning (RL) methods focus on learning optimal policies over low-level action spaces. While these methods can perform well in their training environments, they lack the flexibility to transfer to new tasks. Instead, RL agents that can act over useful, temporally extended skills rather than low-level actions can learn new tasks more easily. Prior work in skill-based RL either requires expert supervision to define useful skills, which is hard to scale, or learns a skill-space from offline data with heuristics that limit the adaptability of the skills, making them difficult to transfer during downstream RL. Our approach, EXTRACT, instead utilizes pre-trained vision language models to extract a discrete set of semantically meaningful skills from offline data, each of which is parameterized by continuous arguments, without human supervision. This skill parameterization allows robots to learn new tasks by only needing to learn when to select a specific skill and how to modify its arguments for the specific task. We demonstrate through experiments in sparse-reward, image-based, robot manipulation environments that EXTRACT can more quickly learn new tasks than prior works, with major gains in sample efficiency and performance over prior skill-based RL. Website at https://www.jessezhang.net/projects/extract/.
title EXTRACT: Efficient Policy Learning by Extracting Transferable Robot Skills from Offline Data
topic Robotics
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2406.17768