Federated Unlearning Model Recovery in Data with Skewed Label Distributions

TL;DR

This paper introduces a novel federated unlearning recovery method tailored for skewed label distributions, using oversampling and denoising techniques to improve model performance after unlearning.

Contribution

It proposes a new recovery approach combining oversampling and density-based denoising to address challenges of biased models in federated unlearning with skewed data.

Findings

Outperforms baseline methods in accuracy on skewed classes

Enhances local dataset completeness for better recovery

Effective in diverse skewness scenarios

Abstract

In federated learning, federated unlearning is a technique that provides clients with a rollback mechanism that allows them to withdraw their data contribution without training from scratch. However, existing research has not considered scenarios with skewed label distributions. Unfortunately, the unlearning of a client with skewed data usually results in biased models and makes it difficult to deliver high-quality service, complicating the recovery process. This paper proposes a recovery method of federated unlearning with skewed label distributions. Specifically, we first adopt a strategy that incorporates oversampling with deep learning to supplement the skewed class data for clients to perform recovery training, therefore enhancing the completeness of their local datasets. Afterward, a density-based denoising method is applied to remove noise from the generated data, further improving the quality of the remaining clients' datasets. Finally, all the remaining clients leverage the enhanced local datasets and engage in iterative training to effectively restore the performance of the unlearning model. Extensive evaluations on commonly used federated learning datasets with varying degrees of skewness show that our method outperforms baseline methods in restoring the performance of the unlearning model, particularly regarding accuracy on the skewed class.