FL Games: A Federated Learning Framework for Distribution Shifts
This addresses the challenge of distribution shifts in federated learning for applications like healthcare or IoT, though it appears incremental as it builds on existing game-theoretic and causal invariance ideas.
The paper tackles the problem of catastrophic generalization in federated learning due to non-i.i.d. client data by proposing FL GAMES, a game-theoretic framework that learns invariant causal features across clients, resulting in high out-of-distribution performance on benchmarks with fewer communication rounds and scalability.
Federated learning aims to train predictive models for data that is distributed across clients, under the orchestration of a server. However, participating clients typically each hold data from a different distribution, which can yield to catastrophic generalization on data from a different client, which represents a new domain. In this work, we argue that in order to generalize better across non-i.i.d. clients, it is imperative to only learn correlations that are stable and invariant across domains. We propose FL GAMES, a game-theoretic framework for federated learning that learns causal features that are invariant across clients. While training to achieve the Nash equilibrium, the traditional best response strategy suffers from high-frequency oscillations. We demonstrate that FL GAMES effectively resolves this challenge and exhibits smooth performance curves. Further, FL GAMES scales well in the number of clients, requires significantly fewer communication rounds, and is agnostic to device heterogeneity. Through empirical evaluation, we demonstrate that FL GAMES achieves high out-of-distribution performance on various benchmarks.