Decentralized Learning Strategies for Estimation Error Minimization with Graph Neural Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Xingran, NaderiAlizadeh, Navid, Ribeiro, Alejandro, Bidokhti, Shirin Saeedi
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908788287602688
author Chen, Xingran
NaderiAlizadeh, Navid
Ribeiro, Alejandro
Bidokhti, Shirin Saeedi
author_facet Chen, Xingran
NaderiAlizadeh, Navid
Ribeiro, Alejandro
Bidokhti, Shirin Saeedi
contents We address real-time sampling and estimation of autoregressive Markovian sources in dynamic yet structurally similar multi-hop wireless networks. Each node caches samples from others and communicates over wireless collision channels, aiming to minimize time-average estimation error via decentralized policies. Due to the high dimensionality of action spaces and complexity of network topologies, deriving optimal policies analytically is intractable. To address this, we propose a graphical multi-agent reinforcement learning framework for policy optimization. Theoretically, we demonstrate that our proposed policies are transferable, allowing a policy trained on one graph to be effectively applied to structurally similar graphs. Numerical experiments demonstrate that (i) our proposed policy outperforms state-of-the-art baselines; (ii) the trained policies are transferable to larger networks, with performance gains increasing with the number of agents; (iii) the graphical training procedure withstands non-stationarity, even when using independent learning techniques; and (iv) recurrence is pivotal in both independent learning and centralized training and decentralized execution, and improves the resilience to non-stationarity.
format Preprint
id arxiv_https___arxiv_org_abs_2601_12662
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Decentralized Learning Strategies for Estimation Error Minimization with Graph Neural Networks
Chen, Xingran
NaderiAlizadeh, Navid
Ribeiro, Alejandro
Bidokhti, Shirin Saeedi
Machine Learning
Signal Processing
We address real-time sampling and estimation of autoregressive Markovian sources in dynamic yet structurally similar multi-hop wireless networks. Each node caches samples from others and communicates over wireless collision channels, aiming to minimize time-average estimation error via decentralized policies. Due to the high dimensionality of action spaces and complexity of network topologies, deriving optimal policies analytically is intractable. To address this, we propose a graphical multi-agent reinforcement learning framework for policy optimization. Theoretically, we demonstrate that our proposed policies are transferable, allowing a policy trained on one graph to be effectively applied to structurally similar graphs. Numerical experiments demonstrate that (i) our proposed policy outperforms state-of-the-art baselines; (ii) the trained policies are transferable to larger networks, with performance gains increasing with the number of agents; (iii) the graphical training procedure withstands non-stationarity, even when using independent learning techniques; and (iv) recurrence is pivotal in both independent learning and centralized training and decentralized execution, and improves the resilience to non-stationarity.
title Decentralized Learning Strategies for Estimation Error Minimization with Graph Neural Networks
topic Machine Learning
Signal Processing
url https://arxiv.org/abs/2601.12662