aiqbench analysis
This research introduces pFedMARL, a framework that uses Multi-Agent Reinforcement Learning to dynamically adjust aggregation strategies in Federated Learning environments with non-IID data distributions. By employing server-side and client-side agents, the method effectively balances global model robustness with local personalization, outperforming standard baselines like FedAvg even under adversarial conditions. The approach demonstrates significant potential for deploying privacy-preserving machine learning systems where data heterogeneity is a critical constraint.
Cooperative Multi-Agent Reinforcement Learning for Adaptive Aggregation in Semi-Supervised Federated Learning with non-IID Data
arXiv:2608.25794v1 Announce Type: new Abstract: Federated Learning (FL) enables distributed training of machine learning models while preserving data privacy. However, FL struggles with heterogeneous, non-IID client data distributions, resulting in sub-optimal and biased global models. In this paper, we propose pFedMARL, a novel approach leveraging Multi-Agent Reinforcement Learning (MARL) with Twin Delayed Deep Deterministic Policy Gradient (TD3) to dynamically adapt aggregation strategies in FL settings. Our method employs a server-side agent adjusting client contributions to optimize global model robustness and client-side agents balancing global and local updates to personalize models effectively without pre-training. We demonstrate superior performance of pFedMARL for training a semi-supervised audio spectrogram transformer, matching or outperforming FedAvg, Ditto, and local training approaches across multiple non-IID scenarios and in the presence of adversarial clients. Our results indicate that pFedMARL actively improves accuracy, robustness, and fairness, making it suitable for real-world deployments.
Abstract:Federated Learning (FL) enables distributed training of machine learning models while preserving data privacy. However, FL struggles with heterogeneous, non-IID client data distributions, resulting in sub-optimal and biased global models. In this paper, we propose pFedMARL, a novel approach leveraging Multi-Agent Reinforcement Learning (MARL) with Twin Delayed Deep Deterministic Policy Gradient (TD3) to dynamically adapt aggregation strategies in FL settings. Our method employs a server-side agent adjusting client contributions to optimize global model robustness and client-side agents balancing global and local updates to personalize models effectively without pre-training. We demonstrate superior performance of pFedMARL for training a semi-supervised audio spectrogram transformer, matching or outperforming FedAvg, Ditto, and local training approaches across multiple non-IID scenarios and in the presence of adversarial clients. Our results indicate that pFedMARL actively improves accuracy, robustness, and fairness, making it suitable for real-world deployments.
Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC); Sound (cs.SD); Audio and Speech Processing (eess.AS); Signal Processing (eess.SP)
arXiv:2608.25794 [cs.LG]
(or arXiv:2608.25794v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2608.25794 arXiv-issued DOI via DataCite (pending registration)
https://doi.org/10.1109/ICASSP55912.2026.11462172 DOI(s) linking to related resources
Full text under the publisher's open licence · 178 words
Related coverage
SIGMA: Structured Noise-Effect-Aware Grouped Multi-Agent Aggregation
arXiv · Aug 28, 2026
Game-Theoretic Disempowerment: Loss of Control to Agencyless AI
LessWrong · Sep 27, 2026
Encoded Coordination on the Open Web
LessWrong · Sep 23, 2026
Cooperative Long Rope Skipping via Multi-Agent Reinforcement Learning
arXiv · Jun 9, 2026