Toward Efficient Data-Free Unlearning
This work addresses data-free unlearning for machine learning systems, but it appears incremental as it builds on prior distillation-based methods.
The paper tackled the challenge of data-free machine unlearning by addressing over-filtering in synthetic samples, which hindered efficient knowledge retention, and proposed the Inhibited Synthetic PostFilter (ISPF) method, which outperformed existing approaches in experiments.
Machine unlearning without access to real data distribution is challenging. The existing method based on data-free distillation achieved unlearning by filtering out synthetic samples containing forgetting information but struggled to distill the retaining-related knowledge efficiently. In this work, we analyze that such a problem is due to over-filtering, which reduces the synthesized retaining-related information. We propose a novel method, Inhibited Synthetic PostFilter (ISPF), to tackle this challenge from two perspectives: First, the Inhibited Synthetic, by reducing the synthesized forgetting information; Second, the PostFilter, by fully utilizing the retaining-related information in synthesized samples. Experimental results demonstrate that the proposed ISPF effectively tackles the challenge and outperforms existing methods.