Private and Robust Contribution Evaluation in Federated Learning

TL;DR

This paper proposes two new contribution evaluation methods for federated learning that ensure fairness, privacy, and robustness, addressing limitations of existing approaches and improving client ranking and model performance.

Contribution

Introduction of two marginal-difference contribution scores compatible with secure aggregation, enhancing fairness, privacy, and robustness in federated learning.

Findings

Scores outperform existing baselines in client ranking

Methods improve downstream model performance

Enhanced detection of malicious participants

Abstract

Cross-silo federated learning allows multiple organizations to collaboratively train machine learning models without sharing raw data, but client updates can still leak sensitive information through inference attacks. Secure aggregation protects privacy by hiding individual updates, yet it complicates contribution evaluation, which is critical for fair rewards and detecting low-quality or malicious participants. Existing marginal-contribution methods, such as the Shapley value, are incompatible with secure aggregation, and practical alternatives, such as Leave-One-Out, are crude and rely on self-evaluation. We introduce two marginal-difference contribution scores compatible with secure aggregation. Fair-Private satisfies standard fairness axioms, while Everybody-Else eliminates self-evaluation and provides resistance to manipulation, addressing a largely overlooked vulnerability. We provide theoretical guarantees for fairness, privacy, robustness, and computational efficiency, and evaluate our methods on multiple medical image datasets and CIFAR10 in cross-silo settings. Our scores consistently outperform existing baselines, better approximate Shapley-induced client rankings, and improve downstream model performance as well as misbehavior detection. These results demonstrate that fairness, privacy, robustness, and practical utility can be achieved jointly in federated contribution evaluation, offering a principled solution for real-world cross-silo deployments.