Machine Unlearning with Minimal Gradient Dependence for High Unlearning Ratios

TL;DR

Mini-Unlearning is a scalable, efficient method for machine unlearning that uses minimal historical gradients and contraction mapping to effectively remove private data traces while maintaining model performance and security.

Contribution

The paper introduces Mini-Unlearning, a novel unlearning approach that reduces reliance on extensive historical gradients through contraction mapping, enabling high unlearning ratios and improved security.

Findings

Outperforms existing unlearning methods in accuracy and security.

Effective at high unlearning ratios with minimal gradient dependence.

Enhances resistance to membership inference attacks.

Abstract

In the context of machine unlearning, the primary challenge lies in effectively removing traces of private data from trained models while maintaining model performance and security against privacy attacks like membership inference attacks. Traditional gradient-based unlearning methods often rely on extensive historical gradients, which becomes impractical with high unlearning ratios and may reduce the effectiveness of unlearning. Addressing these limitations, we introduce Mini-Unlearning, a novel approach that capitalizes on a critical observation: unlearned parameters correlate with retrained parameters through contraction mapping. Our method, Mini-Unlearning, utilizes a minimal subset of historical gradients and leverages this contraction mapping to facilitate scalable, efficient unlearning. This lightweight, scalable method significantly enhances model accuracy and strengthens resistance to membership inference attacks. Our experiments demonstrate that Mini-Unlearning not only works under higher unlearning ratios but also outperforms existing techniques in both accuracy and security, offering a promising solution for applications requiring robust unlearning capabilities.