Towards Aligned Data Forgetting via Twin Machine Unlearning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Niu, Zhenxing, Ji, Haoxuan, Sun, Yuyao, Lin, Zheng, Gao, Fei, Wang, Yuhang, Gao, Haichao
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915117308837888
author Niu, Zhenxing
Ji, Haoxuan
Sun, Yuyao
Lin, Zheng
Gao, Fei
Wang, Yuhang
Gao, Haichao
author_facet Niu, Zhenxing
Ji, Haoxuan
Sun, Yuyao
Lin, Zheng
Gao, Fei
Wang, Yuhang
Gao, Haichao
contents Modern privacy regulations have spurred the evolution of machine unlearning, a technique enabling a trained model to efficiently forget specific training data. In prior unlearning methods, the concept of "data forgetting" is often interpreted and implemented as achieving zero classification accuracy on such data. Nevertheless, the authentic aim of machine unlearning is to achieve alignment between the unlearned model and the gold model, i.e., encouraging them to have identical classification accuracy. On the other hand, the gold model often exhibits non-zero classification accuracy due to its generalization ability. To achieve aligned data forgetting, we propose a Twin Machine Unlearning (TMU) approach, where a twin unlearning problem is defined corresponding to the original unlearning problem. Consequently, the generalization-label predictor trained on the twin problem can be transferred to the original problem, facilitating aligned data forgetting. Comprehensive empirical experiments illustrate that our approach significantly enhances the alignment between the unlearned model and the gold model.
format Preprint
id arxiv_https___arxiv_org_abs_2501_08615
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Towards Aligned Data Forgetting via Twin Machine Unlearning
Niu, Zhenxing
Ji, Haoxuan
Sun, Yuyao
Lin, Zheng
Gao, Fei
Wang, Yuhang
Gao, Haichao
Machine Learning
Modern privacy regulations have spurred the evolution of machine unlearning, a technique enabling a trained model to efficiently forget specific training data. In prior unlearning methods, the concept of "data forgetting" is often interpreted and implemented as achieving zero classification accuracy on such data. Nevertheless, the authentic aim of machine unlearning is to achieve alignment between the unlearned model and the gold model, i.e., encouraging them to have identical classification accuracy. On the other hand, the gold model often exhibits non-zero classification accuracy due to its generalization ability. To achieve aligned data forgetting, we propose a Twin Machine Unlearning (TMU) approach, where a twin unlearning problem is defined corresponding to the original unlearning problem. Consequently, the generalization-label predictor trained on the twin problem can be transferred to the original problem, facilitating aligned data forgetting. Comprehensive empirical experiments illustrate that our approach significantly enhances the alignment between the unlearned model and the gold model.
title Towards Aligned Data Forgetting via Twin Machine Unlearning
topic Machine Learning
url https://arxiv.org/abs/2501.08615