UnlearnShield: Shielding Forgotten Privacy against Unlearning Inversion

TL;DR

UnlearnShield is a novel defense mechanism designed to protect privacy by preventing data reconstruction attacks in machine unlearning, balancing privacy, accuracy, and forgetting effectiveness.

Contribution

It introduces directional perturbations and a constraint module to defend against unlearning inversion without sacrificing model utility.

Findings

Effective in reducing data reconstruction risk

Maintains a good balance between privacy and accuracy

Outperforms existing methods in experiments

Abstract

Machine unlearning is an emerging technique that aims to remove the influence of specific data from trained models, thereby enhancing privacy protection. However, recent research has uncovered critical privacy vulnerabilities, showing that adversaries can exploit unlearning inversion to reconstruct data that was intended to be erased. Despite the severity of this threat, dedicated defenses remain lacking. To address this gap, we propose UnlearnShield, the first defense specifically tailored to counter unlearning inversion. UnlearnShield introduces directional perturbations in the cosine representation space and regulates them through a constraint module to jointly preserve model accuracy and forgetting efficacy, thereby reducing inversion risk while maintaining utility. Experiments demonstrate that it achieves a good trade-off among privacy protection, accuracy, and forgetting.