A New Perspective on Privacy Protection in Federated Learning with Granular-Ball Computing

2025-01-09

AI Summary

To address weak input-data privacy protection in federated learning (FL), this paper proposes Granular-Ball Federated Learning (GrBFL), the first FL framework to incorporate granular-ball computing at the input layer. Methodologically, we design a variance-constrained 2D binary search algorithm for adaptive image segmentation, partitioning raw images into coarse-grained granular-ball regions; a graph structure is then constructed to model local semantic relationships, thereby reducing raw-pixel exposure at the source. Experiments on multiple image classification benchmarks demonstrate that GrBFL surpasses state-of-the-art FL methods across three critical dimensions: privacy protection strength, communication overhead, and model accuracy—achieving synergistic improvement. These results empirically validate the effectiveness of coarse-grained input representation in balancing privacy preservation and model utility.

Technology Category

• Machine Learning: Privacy
• Computer Vision: Bias, Fairness & Privacy
• Search and Optimization: Learning to Search

Application Category

• Security and Privacy: Security and privacy of machine learning and AI applications
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

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.

Problem

• Federated Learning
• Privacy Protection
• Image Recognition

Innovation

• Granularity Sphere Federated Learning
• Optimal Image Partitioning
• Privacy-Preserving Model Efficiency