A Best-of-Both-Worlds Algorithm for Constrained MDPs with Long-Term Constraints

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Germano, Jacopo, Stradi, Francesco Emanuele, Genalti, Gianmarco, Castiglioni, Matteo, Marchesi, Alberto, Gatti, Nicola
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910581050572800
author Germano, Jacopo
Stradi, Francesco Emanuele
Genalti, Gianmarco
Castiglioni, Matteo
Marchesi, Alberto
Gatti, Nicola
author_facet Germano, Jacopo
Stradi, Francesco Emanuele
Genalti, Gianmarco
Castiglioni, Matteo
Marchesi, Alberto
Gatti, Nicola
contents We study online learning in episodic constrained Markov decision processes (CMDPs), where the learner aims at collecting as much reward as possible over the episodes, while satisfying some long-term constraints during the learning process. Rewards and constraints can be selected either stochastically or adversarially, and the transition function is not known to the learner. While online learning in classical (unconstrained) MDPs has received considerable attention over the last years, the setting of CMDPs is still largely unexplored. This is surprising, since in real-world applications, such as, e.g., autonomous driving, automated bidding, and recommender systems, there are usually additional constraints and specifications that an agent has to obey during the learning process. In this paper, we provide the first best-of-both-worlds algorithm for CMDPs with long-term constraints, in the flavor of Balseiro et al. (2023). Our algorithm is capable of handling settings in which rewards and constraints are selected either stochastically or adversarially, without requiring any knowledge of the underling process. Moreover, our algorithm matches state-of-the-art regret and constraint violation bounds for settings in which constraints are selected stochastically, while it is the first to provide guarantees in the case in which they are chosen adversarially.
format Preprint
id arxiv_https___arxiv_org_abs_2304_14326
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle A Best-of-Both-Worlds Algorithm for Constrained MDPs with Long-Term Constraints
Germano, Jacopo
Stradi, Francesco Emanuele
Genalti, Gianmarco
Castiglioni, Matteo
Marchesi, Alberto
Gatti, Nicola
Machine Learning
We study online learning in episodic constrained Markov decision processes (CMDPs), where the learner aims at collecting as much reward as possible over the episodes, while satisfying some long-term constraints during the learning process. Rewards and constraints can be selected either stochastically or adversarially, and the transition function is not known to the learner. While online learning in classical (unconstrained) MDPs has received considerable attention over the last years, the setting of CMDPs is still largely unexplored. This is surprising, since in real-world applications, such as, e.g., autonomous driving, automated bidding, and recommender systems, there are usually additional constraints and specifications that an agent has to obey during the learning process. In this paper, we provide the first best-of-both-worlds algorithm for CMDPs with long-term constraints, in the flavor of Balseiro et al. (2023). Our algorithm is capable of handling settings in which rewards and constraints are selected either stochastically or adversarially, without requiring any knowledge of the underling process. Moreover, our algorithm matches state-of-the-art regret and constraint violation bounds for settings in which constraints are selected stochastically, while it is the first to provide guarantees in the case in which they are chosen adversarially.
title A Best-of-Both-Worlds Algorithm for Constrained MDPs with Long-Term Constraints
topic Machine Learning
url https://arxiv.org/abs/2304.14326