FedRTS: Federated Robust Pruning via Combinatorial Thompson Sampling

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Huang, Hong, Yang, Hai, Chen, Yuan, Ye, Jiaxun, Wu, Dapeng
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908596059504640
author Huang, Hong
Yang, Hai
Chen, Yuan
Ye, Jiaxun
Wu, Dapeng
author_facet Huang, Hong
Yang, Hai
Chen, Yuan
Ye, Jiaxun
Wu, Dapeng
contents 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
format Preprint
id arxiv_https___arxiv_org_abs_2501_19122
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FedRTS: Federated Robust Pruning via Combinatorial Thompson Sampling
Huang, Hong
Yang, Hai
Chen, Yuan
Ye, Jiaxun
Wu, Dapeng
Machine Learning
Artificial Intelligence
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
title FedRTS: Federated Robust Pruning via Combinatorial Thompson Sampling
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2501.19122