Poisoning Semi-supervised Federated Learning via Unlabeled Data: Attacks and Defenses

TL;DR

This paper uncovers a vulnerability in semi-supervised federated learning where malicious unlabeled data can poison the model, and proposes a defense strategy based on client selection and quality aggregation to mitigate such attacks.

Contribution

It introduces a novel poisoning attack exploiting semi-supervised learning properties and proposes a minimax optimization-based client selection defense to enhance robustness.

Findings

Poisoning attack effective with only 0.1% malicious data

Defense strategy significantly reduces attack impact

Attacks and defenses validated on multiple datasets

Abstract

Semi-supervised Federated Learning (SSFL) has recently drawn much attention due to its practical consideration, i.e., the clients may only have unlabeled data. In practice, these SSFL systems implement semi-supervised training by assigning a "guessed" label to the unlabeled data near the labeled data to convert the unsupervised problem into a fully supervised problem. However, the inherent properties of such semi-supervised training techniques create a new attack surface. In this paper, we discover and reveal a simple yet powerful poisoning attack against SSFL. Our attack utilizes the natural characteristic of semi-supervised learning to cause the model to be poisoned by poisoning unlabeled data. Specifically, the adversary just needs to insert a small number of maliciously crafted unlabeled samples (e.g., only 0.1\% of the dataset) to infect model performance and misclassification. Extensive case studies have shown that our attacks are effective on different datasets and common semi-supervised learning methods. To mitigate the attacks, we propose a defense, i.e., a minimax optimization-based client selection strategy, to enable the server to select the clients who hold the correct label information and high-quality updates. Our defense further employs a quality-based aggregation rule to strengthen the contributions of the selected updates. Evaluations under different attack conditions show that the proposed defense can well alleviate such unlabeled poisoning attacks. Our study unveils the vulnerability of SSFL to unlabeled poisoning attacks and provides the community with potential defense methods.