AdaBest: Minimizing Client Drift in Federated Learning via Adaptive Bias Estimation
This work addresses the challenge of model performance degradation in federated learning for decentralized systems, offering a practical improvement over existing methods.
The paper tackles the problem of client drift in federated learning due to data heterogeneity by proposing an adaptive algorithm for accurate drift estimation, which converges faster and achieves higher accuracy than baselines across various benchmarks.
In Federated Learning (FL), a number of clients or devices collaborate to train a model without sharing their data. Models are optimized locally at each client and further communicated to a central hub for aggregation. While FL is an appealing decentralized training paradigm, heterogeneity among data from different clients can cause the local optimization to drift away from the global objective. In order to estimate and therefore remove this drift, variance reduction techniques have been incorporated into FL optimization recently. However, these approaches inaccurately estimate the clients' drift and ultimately fail to remove it properly. In this work, we propose an adaptive algorithm that accurately estimates drift across clients. In comparison to previous works, our approach necessitates less storage and communication bandwidth, as well as lower compute costs. Additionally, our proposed methodology induces stability by constraining the norm of estimates for client drift, making it more practical for large scale FL. Experimental findings demonstrate that the proposed algorithm converges significantly faster and achieves higher accuracy than the baselines across various FL benchmarks.