Guardado en:
Detalles Bibliográficos
Autores principales: Yang, Yan, Gao, Bin, Yuan, Ya-xiang
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2405.19697
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909513005662208
author Yang, Yan
Gao, Bin
Yuan, Ya-xiang
author_facet Yang, Yan
Gao, Bin
Yuan, Ya-xiang
contents Bilevel reinforcement learning (RL), which features intertwined two-level problems, has attracted growing interest recently. The inherent non-convexity of the lower-level RL problem is, however, to be an impediment to developing bilevel optimization methods. By employing the fixed point equation associated with the regularized RL, we characterize the hyper-gradient via fully first-order information, thus circumventing the assumption of lower-level convexity. This, remarkably, distinguishes our development of hyper-gradient from the general AID-based bilevel frameworks since we take advantage of the specific structure of RL problems. Moreover, we design both model-based and model-free bilevel reinforcement learning algorithms, facilitated by access to the fully first-order hyper-gradient. Both algorithms enjoy the convergence rate $O(ε^{-1})$. To extend the applicability, a stochastic version of the model-free algorithm is proposed, along with results on its iteration and sample complexity. In addition, numerical experiments demonstrate that the hyper-gradient indeed serves as an integration of exploitation and exploration.
format Preprint
id arxiv_https___arxiv_org_abs_2405_19697
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Bilevel reinforcement learning via the development of hyper-gradient without lower-level convexity
Yang, Yan
Gao, Bin
Yuan, Ya-xiang
Optimization and Control
Artificial Intelligence
Machine Learning
Bilevel reinforcement learning (RL), which features intertwined two-level problems, has attracted growing interest recently. The inherent non-convexity of the lower-level RL problem is, however, to be an impediment to developing bilevel optimization methods. By employing the fixed point equation associated with the regularized RL, we characterize the hyper-gradient via fully first-order information, thus circumventing the assumption of lower-level convexity. This, remarkably, distinguishes our development of hyper-gradient from the general AID-based bilevel frameworks since we take advantage of the specific structure of RL problems. Moreover, we design both model-based and model-free bilevel reinforcement learning algorithms, facilitated by access to the fully first-order hyper-gradient. Both algorithms enjoy the convergence rate $O(ε^{-1})$. To extend the applicability, a stochastic version of the model-free algorithm is proposed, along with results on its iteration and sample complexity. In addition, numerical experiments demonstrate that the hyper-gradient indeed serves as an integration of exploitation and exploration.
title Bilevel reinforcement learning via the development of hyper-gradient without lower-level convexity
topic Optimization and Control
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2405.19697