Off-Policy Temporal Difference Learning for Perturbed Markov Decision Processes: Theoretical Insights and Extensive Simulations

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Forootani, Ali, Iervolino, Raffaele, Tipaldi, Massimo, Khosravi, Mohammad
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917944411291648
author Forootani, Ali
Iervolino, Raffaele
Tipaldi, Massimo
Khosravi, Mohammad
author_facet Forootani, Ali
Iervolino, Raffaele
Tipaldi, Massimo
Khosravi, Mohammad
contents Dynamic Programming suffers from the curse of dimensionality due to large state and action spaces, a challenge further compounded by uncertainties in the environment. To mitigate these issue, we explore an off-policy based Temporal Difference Approximate Dynamic Programming approach that preserves contraction mapping when projecting the problem into a subspace of selected features, accounting for the probability distribution of the perturbed transition probability matrix. We further demonstrate how this Approximate Dynamic Programming approach can be implemented as a particular variant of the Temporal Difference learning algorithm, adapted for handling perturbations. To validate our theoretical findings, we provide a numerical example using a Markov Decision Process corresponding to a resource allocation problem.
format Preprint
id arxiv_https___arxiv_org_abs_2502_18415
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Off-Policy Temporal Difference Learning for Perturbed Markov Decision Processes: Theoretical Insights and Extensive Simulations
Forootani, Ali
Iervolino, Raffaele
Tipaldi, Massimo
Khosravi, Mohammad
Systems and Control
Dynamic Programming suffers from the curse of dimensionality due to large state and action spaces, a challenge further compounded by uncertainties in the environment. To mitigate these issue, we explore an off-policy based Temporal Difference Approximate Dynamic Programming approach that preserves contraction mapping when projecting the problem into a subspace of selected features, accounting for the probability distribution of the perturbed transition probability matrix. We further demonstrate how this Approximate Dynamic Programming approach can be implemented as a particular variant of the Temporal Difference learning algorithm, adapted for handling perturbations. To validate our theoretical findings, we provide a numerical example using a Markov Decision Process corresponding to a resource allocation problem.
title Off-Policy Temporal Difference Learning for Perturbed Markov Decision Processes: Theoretical Insights and Extensive Simulations
topic Systems and Control
url https://arxiv.org/abs/2502.18415