EVE: Efficient Verification of Data Erasure through Customized Perturbation in Approximate Unlearning

2026-02-03

AI Summary

This work addresses the challenge of efficiently verifying machine unlearning without access to the model’s initial training process. The authors propose a novel verification method that requires neither backdoor insertion nor any intervention during training. By constructing tailored perturbations that induce detectable differences in model predictions before and after unlearning, the approach formulates perturbation generation as an adversarial optimization problem that aligns the gradient of the unlearning objective with the gradient of decision boundary shifts. This formulation enables, for the first time, a fully decoupled and efficient verification mechanism independent of the original training procedure. Experimental results demonstrate that the proposed method significantly outperforms existing techniques in both accuracy and computational efficiency, offering a practical and reliable solution for validating data removal in machine learning models.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Adversarial Attacks & Robustness
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Verifying whether the machine unlearning process has been properly executed is critical but remains underexplored. Some existing approaches propose unlearning verification methods based on backdooring techniques. However, these methods typically require participation in the model's initial training phase to backdoor the model for later verification, which is inefficient and impractical. In this paper, we propose an efficient verification of erasure method (EVE) for verifying machine unlearning without requiring involvement in the model's initial training process. The core idea is to perturb the unlearning data to ensure the model prediction of the specified samples will change before and after unlearning with perturbed data. The unlearning users can leverage the observation of the changes as a verification signal. Specifically, the perturbations are designed with two key objectives: ensuring the unlearning effect and altering the unlearned model's prediction of target samples. We formalize the perturbation generation as an adversarial optimization problem, solving it by aligning the unlearning gradient with the gradient of boundary change for target samples. We conducted extensive experiments, and the results show that EVE can verify machine unlearning without involving the model's initial training process, unlike backdoor-based methods. Moreover, EVE significantly outperforms state-of-the-art unlearning verification methods, offering significant speedup in efficiency while enhancing verification accuracy. The source code of EVE is released at \uline{https://anonymous.4open.science/r/EVE-C143}, providing a novel tool for verification of machine unlearning.

Problem

• machine unlearning
• unlearning verification
• data erasure
• efficient verification
• model verification

Innovation

• machine unlearning verification
• customized perturbation
• adversarial optimization
• data erasure
• gradient alignment