Saved in:
Bibliographic Details
Main Authors: Lim, Han-Dong, Lee, Donghwan
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2310.00638
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910939274543104
author Lim, Han-Dong
Lee, Donghwan
author_facet Lim, Han-Dong
Lee, Donghwan
contents The goal of this paper is to investigate distributed temporal difference (TD) learning for a networked multi-agent Markov decision process. The proposed approach is based on distributed optimization algorithms, which can be interpreted as primal-dual Ordinary differential equation (ODE) dynamics subject to null-space constraints. Based on the exponential convergence behavior of the primal-dual ODE dynamics subject to null-space constraints, we examine the behavior of the final iterate in various distributed TD-learning scenarios, considering both constant and diminishing step-sizes and incorporating both i.i.d. and Markovian observation models. Unlike existing methods, the proposed algorithm does not require the assumption that the underlying communication network structure is characterized by a doubly stochastic matrix.
format Preprint
id arxiv_https___arxiv_org_abs_2310_00638
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle A primal-dual perspective for distributed TD-learning
Lim, Han-Dong
Lee, Donghwan
Machine Learning
Optimization and Control
The goal of this paper is to investigate distributed temporal difference (TD) learning for a networked multi-agent Markov decision process. The proposed approach is based on distributed optimization algorithms, which can be interpreted as primal-dual Ordinary differential equation (ODE) dynamics subject to null-space constraints. Based on the exponential convergence behavior of the primal-dual ODE dynamics subject to null-space constraints, we examine the behavior of the final iterate in various distributed TD-learning scenarios, considering both constant and diminishing step-sizes and incorporating both i.i.d. and Markovian observation models. Unlike existing methods, the proposed algorithm does not require the assumption that the underlying communication network structure is characterized by a doubly stochastic matrix.
title A primal-dual perspective for distributed TD-learning
topic Machine Learning
Optimization and Control
url https://arxiv.org/abs/2310.00638