Federated Learning for Face Recognition with Gradient Correction

TL;DR

This paper introduces FedGC, a federated learning framework for face recognition that enhances privacy by correcting gradients through a novel regularizer, achieving performance comparable to centralized methods.

Contribution

The paper proposes a gradient correction method with a softmax-based regularizer for privacy-preserving federated face recognition, ensuring higher privacy and comparable accuracy.

Findings

FedGC matches centralized method performance on benchmarks.

Gradient correction improves privacy without sacrificing accuracy.

Theoretically validated as a proper loss function.

Abstract

With increasing appealing to privacy issues in face recognition, federated learning has emerged as one of the most prevalent approaches to study the unconstrained face recognition problem with private decentralized data. However, conventional decentralized federated algorithm sharing whole parameters of networks among clients suffers from privacy leakage in face recognition scene. In this work, we introduce a framework, FedGC, to tackle federated learning for face recognition and guarantees higher privacy. We explore a novel idea of correcting gradients from the perspective of backward propagation and propose a softmax-based regularizer to correct gradients of class embeddings by precisely injecting a cross-client gradient term. Theoretically, we show that FedGC constitutes a valid loss function similar to standard softmax. Extensive experiments have been conducted to validate the superiority of FedGC which can match the performance of conventional centralized methods utilizing full training dataset on several popular benchmark datasets.