Factor Decorrelation Enhanced Data Removal from Deep Predictive Models
This addresses privacy protection and regulatory compliance needs for users and organizations by improving data removal in models, though it appears incremental as it builds on existing removal techniques.
The paper tackles the problem of sensitive data removal in deep predictive models, which often causes distributional shifts that harm performance, especially in out-of-distribution scenarios. The proposed approach uses factor decorrelation and loss perturbation to achieve high predictive accuracy and robustness, outperforming baselines on five benchmark datasets.
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.