TurboSVM-FL: Boosting Federated Learning through SVM Aggregation for Lazy Clients
Mengdi Wang, Anna Bodonhelyi, Efe Bozkir, Enkelejda Kasneci
TL;DR
TurboSVM-FL tackles the slow convergence of cross-device federated learning under data heterogeneity by introducing a server-side aggregation strategy that leverages support vector machines on class embeddings. It performs selective aggregation by focusing on SVM support vectors and enforces max-margin spread-out regularization to maintain discriminative class representations, all without adding client-side computation. Experimental results on FEMNIST, CelebA, and Shakespeare show substantial reductions in communication rounds and improved test metrics for FEMNIST and CelebA, with Shakespeare remaining competitive against adaptive baselines. The method is particularly advantageous for edge devices with limited resources and can complement existing FL strategies, with kernelization and multi-task extensions highlighted as promising future directions.
Abstract
Federated learning is a distributed collaborative machine learning paradigm that has gained strong momentum in recent years. In federated learning, a central server periodically coordinates models with clients and aggregates the models trained locally by clients without necessitating access to local data. Despite its potential, the implementation of federated learning continues to encounter several challenges, predominantly the slow convergence that is largely due to data heterogeneity. The slow convergence becomes particularly problematic in cross-device federated learning scenarios where clients may be strongly limited by computing power and storage space, and hence counteracting methods that induce additional computation or memory cost on the client side such as auxiliary objective terms and larger training iterations can be impractical. In this paper, we propose a novel federated aggregation strategy, TurboSVM-FL, that poses no additional computation burden on the client side and can significantly accelerate convergence for federated classification task, especially when clients are "lazy" and train their models solely for few epochs for next global aggregation. TurboSVM-FL extensively utilizes support vector machine to conduct selective aggregation and max-margin spread-out regularization on class embeddings. We evaluate TurboSVM-FL on multiple datasets including FEMNIST, CelebA, and Shakespeare using user-independent validation with non-iid data distribution. Our results show that TurboSVM-FL can significantly outperform existing popular algorithms on convergence rate and reduce communication rounds while delivering better test metrics including accuracy, F1 score, and MCC.