Open Problem: Order Optimal Regret Bounds for Kernel-Based Reinforcement Learning

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autor principal: Vakili, Sattar
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910497696120832
author Vakili, Sattar
author_facet Vakili, Sattar
contents Reinforcement Learning (RL) has shown great empirical success in various application domains. The theoretical aspects of the problem have been extensively studied over past decades, particularly under tabular and linear Markov Decision Process structures. Recently, non-linear function approximation using kernel-based prediction has gained traction. This approach is particularly interesting as it naturally extends the linear structure, and helps explain the behavior of neural-network-based models at their infinite width limit. The analytical results however do not adequately address the performance guarantees for this case. We will highlight this open problem, overview existing partial results, and discuss related challenges.
format Preprint
id arxiv_https___arxiv_org_abs_2406_15250
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Open Problem: Order Optimal Regret Bounds for Kernel-Based Reinforcement Learning
Vakili, Sattar
Machine Learning
Reinforcement Learning (RL) has shown great empirical success in various application domains. The theoretical aspects of the problem have been extensively studied over past decades, particularly under tabular and linear Markov Decision Process structures. Recently, non-linear function approximation using kernel-based prediction has gained traction. This approach is particularly interesting as it naturally extends the linear structure, and helps explain the behavior of neural-network-based models at their infinite width limit. The analytical results however do not adequately address the performance guarantees for this case. We will highlight this open problem, overview existing partial results, and discuss related challenges.
title Open Problem: Order Optimal Regret Bounds for Kernel-Based Reinforcement Learning
topic Machine Learning
url https://arxiv.org/abs/2406.15250