Sybil-based Virtual Data Poisoning Attacks in Federated Learning

TL;DR

This paper introduces a novel sybil-based virtual data poisoning attack in federated learning, using gradient matching to reduce computational costs and multiple schemes for target model acquisition, demonstrating superior effectiveness in simulations.

Contribution

It presents a new attack method leveraging sybil nodes and virtual data generation to enhance poisoning effectiveness while lowering computational complexity.

Findings

Outperforms existing attack algorithms in simulations.

Effective under non-i.i.d. data distributions.

Applicable to various federated learning scenarios.

Abstract

Federated learning is vulnerable to poisoning attacks by malicious adversaries. Existing methods often involve high costs to achieve effective attacks. To address this challenge, we propose a sybil-based virtual data poisoning attack, where a malicious client generates sybil nodes to amplify the poisoning model's impact. To reduce neural network computational complexity, we develop a virtual data generation method based on gradient matching. We also design three schemes for target model acquisition, applicable to online local, online global, and offline scenarios. In simulation, our method outperforms other attack algorithms since our method can obtain a global target model under non-independent uniformly distributed data.