FL Games: A federated learning framework for distribution shifts
This addresses the challenge of non-i.i.d. data in federated learning for applications like distributed devices, though it appears incremental as it builds on existing game-theoretic and invariant learning ideas.
The paper tackles the problem of catastrophic failure of federated learning models on unseen domains due to distribution shifts across clients, proposing FL Games, a game-theoretic framework that learns invariant causal features, 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, whereby predictive models with strong in-distribution generalization can fail catastrophically on unseen domains. 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 for learning 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.