Machine Unlearning with Minimal Gradient Dependence for High Unlearning Ratios
This work addresses privacy concerns in machine learning by enabling more efficient and secure unlearning, though it appears incremental as it builds on existing gradient-based methods with a novel optimization.
The paper tackles the challenge of removing private data from trained models in machine unlearning, particularly at high unlearning ratios, by introducing Mini-Unlearning, which uses minimal historical gradients and a contraction mapping approach to improve accuracy and security against membership inference attacks.
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.