Federated Learning Meets Multi-objective Optimization
This addresses fairness and robustness issues in federated learning for edge device users, representing an incremental improvement with a novel method for a known bottleneck.
The paper tackles federated learning by formulating it as multi-objective optimization to ensure fairness and robustness, proposing FedMGDA+, which converges to Pareto stationary solutions and performs favorably against state-of-the-art methods in experiments.
Federated learning has emerged as a promising, massively distributed way to train a joint deep model over large amounts of edge devices while keeping private user data strictly on device. In this work, motivated from ensuring fairness among users and robustness against malicious adversaries, we formulate federated learning as multi-objective optimization and propose a new algorithm FedMGDA+ that is guaranteed to converge to Pareto stationary solutions. FedMGDA+ is simple to implement, has fewer hyperparameters to tune, and refrains from sacrificing the performance of any participating user. We establish the convergence properties of FedMGDA+ and point out its connections to existing approaches. Extensive experiments on a variety of datasets confirm that FedMGDA+ compares favorably against state-of-the-art.