Bandit-based Communication-Efficient Client Selection Strategies for Federated Learning
This work is significant for federated learning practitioners facing communication constraints and intermittent client availability, offering a method to improve convergence speed and fairness without increasing communication overhead.
This paper addresses communication-efficient client selection in federated learning, where only a subset of clients participate in each round. It proposes a bandit-based strategy, UCB-CS, that achieves faster convergence with lower communication overhead compared to prior biased selection methods that either incur additional communication costs or use stale information.
Due to communication constraints and intermittent client availability in federated learning, only a subset of clients can participate in each training round. While most prior works assume uniform and unbiased client selection, recent work on biased client selection has shown that selecting clients with higher local losses can improve error convergence speed. However, previously proposed biased selection strategies either require additional communication cost for evaluating the exact local loss or utilize stale local loss, which can even make the model diverge. In this paper, we present a bandit-based communication-efficient client selection strategy UCB-CS that achieves faster convergence with lower communication overhead. We also demonstrate how client selection can be used to improve fairness.