Attack-Resistant Federated Learning with Residual-based Reweighting

2019/12/24 by Shuhao Fu, Fu, Shuhao, Chulin Xie +5 · 2 citations
Computer Science · #Privacy-Preserving Technologies in Data #Adversarial Robustness in Machine Learning #Cryptography and Data Security

paper · pdf · doi:10.48550/arxiv.1912.11464

Abstract

Federated learning has a variety of applications in multiple domains by utilizing private training data stored on different devices. However, the aggregation process in federated learning is highly vulnerable to adversarial attacks so that the global model may behave abnormally under attacks. To tackle this challenge, we present a novel aggregation algorithm with residual-based reweighting to defend federated learning. Our aggregation algorithm combines repeated median regression with the reweighting scheme in iteratively reweighted least squares. Our experiments show that our aggregation algorithm outperforms other alternative algorithms in the presence of label-flipping and backdoor attacks. We also provide theoretical analysis for our aggregation algorithm.

Citations

• Gradient-based learning applied to document recognition

Cited by

• GShield: Mitigating Poisoning Attacks in Federated Learning
• Poison Once, Refuse Forever: Weaponizing Alignment for Injecting Bias in LLMs

Related

• Robust Federated Learning with Attack-Adaptive Aggregation
• DFedReweighting: A Unified Framework for Objective-Oriented Reweighting in Decentralized Federated Learning
• FedGreed: A Byzantine-Robust Loss-Based Aggregation Method for Federated Learning
• FedRDF: A Robust and Dynamic Aggregation Function against Poisoning Attacks in Federated Learning
• RobustFed: A Truth Inference Approach for Robust Federated Learning