OptiGradTrust: Byzantine-Robust Federated Learning with Multi-Feature Gradient Analysis and Reinforcement Learning-Based Trust Weighting

TL;DR

OptiGradTrust introduces a multi-feature gradient analysis and reinforcement learning-based trust weighting to enhance Byzantine robustness in federated learning, especially under data heterogeneity and attack scenarios.

Contribution

The paper proposes a novel six-dimensional gradient fingerprint and a hybrid RL-attention trust module, along with FedBN-Prox for improved convergence in heterogeneous federated learning environments.

Findings

Achieves up to +1.6% accuracy over FLGuard under non-IID data.

Demonstrates robustness against various Byzantine attack patterns.

Effective in medical imaging and standard datasets.

Abstract

Federated Learning (FL) enables collaborative model training across distributed medical institutions while preserving patient privacy, but remains vulnerable to Byzantine attacks and statistical heterogeneity. We present OptiGradTrust, a comprehensive defense framework that evaluates gradient updates through a novel six-dimensional fingerprint including VAE reconstruction error, cosine similarity metrics, $L_2$ norm, sign-consistency ratio, and Monte Carlo Shapley value, which drive a hybrid RL-attention module for adaptive trust scoring. To address convergence challenges under data heterogeneity, we develop FedBN-Prox (FedBN-P), combining Federated Batch Normalization with proximal regularization for optimal accuracy-convergence trade-offs. Extensive evaluation across MNIST, CIFAR-10, and Alzheimer's MRI datasets under various Byzantine attack scenarios demonstrates significant improvements over state-of-the-art defenses, achieving up to +1.6 percentage points over FLGuard under non-IID conditions while maintaining robust performance against diverse attack patterns through our adaptive learning approach.