FedRTS: Federated Robust Pruning via Combinatorial Thompson Sampling

TL;DR

FedRTS introduces a novel federated pruning framework using Thompson Sampling to create robust, efficient sparse models that perform well under data heterogeneity and limited client participation.

Contribution

The paper proposes FedRTS, a new probabilistic pruning method that improves robustness and efficiency in federated learning over existing greedy approaches.

Findings

Achieves state-of-the-art results in vision and NLP tasks.

Reduces communication costs significantly.

Performs well with heterogeneous data and partial client participation.

Abstract

Federated Learning (FL) enables collaborative model training across distributed clients without data sharing, but its high computational and communication demands strain resource-constrained devices. While existing methods use dynamic pruning to improve efficiency by periodically adjusting sparse model topologies while maintaining sparsity, these approaches suffer from issues such as greedy adjustments, unstable topologies, and communication inefficiency, resulting in less robust models and suboptimal performance under data heterogeneity and partial client availability. To address these challenges, we propose Federated Robust pruning via combinatorial Thompson Sampling (FedRTS), a novel framework designed to develop robust sparse models. FedRTS enhances robustness and performance through its Thompson Sampling-based Adjustment (TSAdj) mechanism, which uses probabilistic decisions informed by stable, farsighted information instead of deterministic decisions reliant on unstable and myopic information in previous methods. Extensive experiments demonstrate that FedRTS achieves state-of-the-art performance in computer vision and natural language processing tasks while reducing communication costs, particularly excelling in scenarios with heterogeneous data distributions and partial client participation. Our codes are available at: https://github.com/Little0o0/FedRTS