Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces pFedMARL, a novel approach that employs Multi-Agent Reinforcement Learning (MARL) with Twin Delayed Deep Deterministic Policy Gradient (TD3) to enhance aggregation strategies in Federated Learning (FL) under non-IID data distributions. By utilizing a server-side agent to optimize client contributions and client-side agents to balance global and local updates, pFedMARL effectively personalizes models without pre-training. The method outperforms existing approaches like FedAvg and Ditto in various non-IID scenarios, demonstrating significant improvements in accuracy, robustness, and fairness for real-world applications.
pFedMARL not only outperforms traditional FL methods but also adapts dynamically to client data heterogeneity, enhancing model robustness and personalization.
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.