acceptodds
Desk rejected at ICLR 2027

Hybrid Framework for Hyperparameter Optimization in Federated Learning.

Desk rejected at ICLR 2027, so left out of the paper lists and the map.

Abstract

Federated Learning (FL) enables collaborative model training across decentralized clients while preserving data privacy, yet optimizing its hyperparameters remains a critical challenge due to data heterogeneity, dynamic resource constraints, and heavy communication overheads. Existing hyperparameter optimization techniques often fail to jointly tune system-level variables, execution dynamics, and internal algorithm parameters. In this paper, we propose PSO-REALF, a novel three-tier hybrid optimization framework designed to maximize efficiency and model accuracy in non-IID federated environments. The architecture leverages a Particle Swarm Optimization (PSO) at the intermediate layer to dynamically adjust core FL hyperparameters, and Resource-Efficient Accelerating Federated Learning (REAFL) at the execution layer to optimize local iteration assignments () under fixed time budgets. Evaluated on non-IID partitions of the MNIST and CIFAR-10 datasets using Convolutional Neural Networks, PSO-REALF demonstrates superior convergence speed and accuracy compared to state-of-the-art baselines like FedEx and Auto-FedRL.

Primary area:
optimization
Keywords:
Federated Learning, Hyperparameter Optimization, Particle Swarm Optimization, REAFL, Deep Learning.

Related papers

Discussion (0)

Sign in to comment.