Work

Toward Efficient Data-Free Unlearning

Year: 2025

Type: conference-paper

Abstract: 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.

Source: Proceedings of the AAAI Conference on Artificial Intelligence

DOI: 10.1609/aaai.v39i21.34393

Authors: Chenhao Zhang, Shaofei Shen, Weitong Chen, Miao Xu

Institutions: The University of Queensland, The University of Adelaide

Language: English

Topic: Privacy-Preserving Technologies in Data

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: diamond

Funder: Australian Research Council

Awards: Towards knowledge discovery from imperfect and evolving data, Detecting Key Concepts from Low-Quality Data for Better Decision