Saved in:
Bibliographic Details
Main Authors: Yang, Wenhao, Li, Lin, Tao, Xiaohui, Shi, Kaize
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.23443
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911181778714624
author Yang, Wenhao
Li, Lin
Tao, Xiaohui
Shi, Kaize
author_facet Yang, Wenhao
Li, Lin
Tao, Xiaohui
Shi, Kaize
contents The imperative of user privacy protection and regulatory compliance necessitates sensitive data removal in model training, yet this process often induces distributional shifts that undermine model performance-particularly in out-of-distribution (OOD) scenarios. We propose a novel data removal approach that enhances deep predictive models through factor decorrelation and loss perturbation. Our approach introduces: (1) a discriminative-preserving factor decorrelation module employing dynamic adaptive weight adjustment and iterative representation updating to reduce feature redundancy and minimize inter-feature correlations. (2) a smoothed data removal mechanism with loss perturbation that creates information-theoretic safeguards against data leakage during removal operations. Extensive experiments on five benchmark datasets show that our approach outperforms other baselines and consistently achieves high predictive accuracy and robustness even under significant distribution shifts. The results highlight its superior efficiency and adaptability in both in-distribution and out-of-distribution scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2509_23443
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Factor Decorrelation Enhanced Data Removal from Deep Predictive Models
Yang, Wenhao
Li, Lin
Tao, Xiaohui
Shi, Kaize
Machine Learning
Artificial Intelligence
The imperative of user privacy protection and regulatory compliance necessitates sensitive data removal in model training, yet this process often induces distributional shifts that undermine model performance-particularly in out-of-distribution (OOD) scenarios. We propose a novel data removal approach that enhances deep predictive models through factor decorrelation and loss perturbation. Our approach introduces: (1) a discriminative-preserving factor decorrelation module employing dynamic adaptive weight adjustment and iterative representation updating to reduce feature redundancy and minimize inter-feature correlations. (2) a smoothed data removal mechanism with loss perturbation that creates information-theoretic safeguards against data leakage during removal operations. Extensive experiments on five benchmark datasets show that our approach outperforms other baselines and consistently achieves high predictive accuracy and robustness even under significant distribution shifts. The results highlight its superior efficiency and adaptability in both in-distribution and out-of-distribution scenarios.
title Factor Decorrelation Enhanced Data Removal from Deep Predictive Models
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2509.23443