Performance Improvement Bounds for Lipschitz Configurable Markov Decision Processes

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Metelli, Alberto Maria
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914688076349440
author Metelli, Alberto Maria
author_facet Metelli, Alberto Maria
contents Configurable Markov Decision Processes (Conf-MDPs) have recently been introduced as an extension of the traditional Markov Decision Processes (MDPs) to model the real-world scenarios in which there is the possibility to intervene in the environment in order to configure some of its parameters. In this paper, we focus on a particular subclass of Conf-MDP that satisfies regularity conditions, namely Lipschitz continuity. We start by providing a bound on the Wasserstein distance between $γ$-discounted stationary distributions induced by changing policy and configuration. This result generalizes the already existing bounds both for Conf-MDPs and traditional MDPs. Then, we derive a novel performance improvement lower bound.
format Preprint
id arxiv_https___arxiv_org_abs_2402_13821
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Performance Improvement Bounds for Lipschitz Configurable Markov Decision Processes
Metelli, Alberto Maria
Machine Learning
Configurable Markov Decision Processes (Conf-MDPs) have recently been introduced as an extension of the traditional Markov Decision Processes (MDPs) to model the real-world scenarios in which there is the possibility to intervene in the environment in order to configure some of its parameters. In this paper, we focus on a particular subclass of Conf-MDP that satisfies regularity conditions, namely Lipschitz continuity. We start by providing a bound on the Wasserstein distance between $γ$-discounted stationary distributions induced by changing policy and configuration. This result generalizes the already existing bounds both for Conf-MDPs and traditional MDPs. Then, we derive a novel performance improvement lower bound.
title Performance Improvement Bounds for Lipschitz Configurable Markov Decision Processes
topic Machine Learning
url https://arxiv.org/abs/2402.13821