Privacy-Preserving Debiasing using Data Augmentation and Machine Unlearning

arXiv:2404.13194v1 Announce Type: new
Abstract: Data augmentation is widely used to mitigate data bias in the training dataset. However, data augmentation exposes machine learning models to privacy attacks, such as membership inference attacks. In this paper, we propose an effective combination of data augmentation and machine unlearning, which can reduce data bias while providing a provable defense against known attacks. Specifically, we maintain the fairness of the trained model with diffusion-based data augmentation, and then utilize multi-shard unlearning to remove …

abstract arxiv attacks augmentation bias combination cs.ai cs.cv cs.lg data data bias dataset however inference machine machine learning machine learning models paper privacy reduce training type unlearning