Syllabus: Portable Curricula for Reinforcement Learning Agents

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Sullivan, Ryan, Pégoud, Ryan, Rehman, Ameen Ur, Yang, Xinchen, Huang, Junyun, Verma, Aayush, Mitra, Nistha, Dickerson, John P.
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866918110761582592
author Sullivan, Ryan
Pégoud, Ryan
Rehman, Ameen Ur
Yang, Xinchen
Huang, Junyun
Verma, Aayush
Mitra, Nistha
Dickerson, John P.
author_facet Sullivan, Ryan
Pégoud, Ryan
Rehman, Ameen Ur
Yang, Xinchen
Huang, Junyun
Verma, Aayush
Mitra, Nistha
Dickerson, John P.
contents Curriculum learning has been a quiet, yet crucial component of many high-profile successes of reinforcement learning. Despite this, it is still a niche topic that is not directly supported by any of the major reinforcement learning libraries. These methods can improve the capabilities and generalization of RL agents, but often require complex changes to training code. We introduce Syllabus, a portable curriculum learning library, as a solution to this problem. Syllabus provides a universal API for curriculum learning, modular implementations of popular automatic curriculum learning methods, and infrastructure that allows them to be easily integrated with asynchronous training code in nearly any RL library. Syllabus provides a minimal API for core curriculum learning components, making it easier to design new algorithms and adapt existing ones to new environments. We demonstrate this by evaluating the algorithms in Syllabus on several new environments, each using agents written in a different RL library. We present the first examples of automatic curriculum learning in NetHack and Neural MMO, two of the most challenging RL benchmarks, and find evidence that existing methods do not directly transfer to complex new environments. Syllabus can be found at https://github.com/RyanNavillus/Syllabus.
format Preprint
id arxiv_https___arxiv_org_abs_2411_11318
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Syllabus: Portable Curricula for Reinforcement Learning Agents
Sullivan, Ryan
Pégoud, Ryan
Rehman, Ameen Ur
Yang, Xinchen
Huang, Junyun
Verma, Aayush
Mitra, Nistha
Dickerson, John P.
Artificial Intelligence
Curriculum learning has been a quiet, yet crucial component of many high-profile successes of reinforcement learning. Despite this, it is still a niche topic that is not directly supported by any of the major reinforcement learning libraries. These methods can improve the capabilities and generalization of RL agents, but often require complex changes to training code. We introduce Syllabus, a portable curriculum learning library, as a solution to this problem. Syllabus provides a universal API for curriculum learning, modular implementations of popular automatic curriculum learning methods, and infrastructure that allows them to be easily integrated with asynchronous training code in nearly any RL library. Syllabus provides a minimal API for core curriculum learning components, making it easier to design new algorithms and adapt existing ones to new environments. We demonstrate this by evaluating the algorithms in Syllabus on several new environments, each using agents written in a different RL library. We present the first examples of automatic curriculum learning in NetHack and Neural MMO, two of the most challenging RL benchmarks, and find evidence that existing methods do not directly transfer to complex new environments. Syllabus can be found at https://github.com/RyanNavillus/Syllabus.
title Syllabus: Portable Curricula for Reinforcement Learning Agents
topic Artificial Intelligence
url https://arxiv.org/abs/2411.11318