通过粗粒度图像分块重构图结构，提升联邦学习隐私与效率
A New Perspective on Privacy Protection in Federated Learning with Granular-Ball Computing
- 将图像按方差约束分块为粗粒度区域，构建图结构输入
- 在保持分类性能前提下，隐私保护更强且训练效率更高
- 适合关注数据隐私的图像联邦学习场景
联邦学习（FL）可在不共享原始数据的前提下实现模型协同训练，但现有方法多聚焦于模型参数与输出层面的隐私保护，忽视了输入层的防护。为此，本文提出一种新型框架——粒球联邦学习（GrBFL），用于图像分类任务。不同于依赖细粒度输入的传统方法，GrBFL将图像分割为具有最优粗粒度的多个区域，并将其重构为图结构。设计了一种基于方差约束的二维二分搜索分割算法，有效去除冗余信息，同时保留关键代表性特征。理论分析与大量实验表明，GrBFL在保障隐私、提升效率的同时维持强健的模型性能，显著优于当前主流联邦学习方法。代码已开源：https://github.com/AIGNLAI/GrBFL。
原文摘要 · Abstract (English)
Federated Learning (FL) facilitates collaborative model training while prioritizing privacy by avoiding direct data sharing. However, most existing articles attempt to address challenges within the model's internal parameters and corresponding outputs, while neglecting to solve them at the input level. To address this gap, we propose a novel framework called Granular-Ball Federated Learning (GrBFL) for image classification. GrBFL diverges from traditional methods that rely on the finest-grained input data. Instead, it segments images into multiple regions with optimal coarse granularity, which are then reconstructed into a graph structure. We designed a two-dimensional binary search segmentation algorithm based on variance constraints for GrBFL, which effectively removes redundant information while preserving key representative features. Extensive theoretical analysis and experiments demonstrate that GrBFL not only safeguards privacy and enhances efficiency but also maintains robust utility, consistently outperforming other state-of-the-art FL methods. The code is available at https://github.com/AIGNLAI/GrBFL.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。