GENIU: A Restricted Data Access Unlearning for Imbalanced Data

TL;DR

GENIU is a novel framework that enables class unlearning in imbalanced datasets with restricted data access by using a VAE-based proxy generator, improving unlearning accuracy without original data.

Contribution

It introduces GENIU, the first practical method for class unlearning in imbalanced data scenarios with limited data access, utilizing a VAE for proxy generation and in-batch tuning.

Findings

GENIU outperforms existing unlearning methods in imbalanced data settings.

The VAE-based proxy generator effectively represents class distributions.

In-batch tuning enhances unlearning performance for majority classes.

Abstract

With the increasing emphasis on data privacy, the significance of machine unlearning has grown substantially. Class unlearning, which involves enabling a trained model to forget data belonging to a specific class learned before, is important as classification tasks account for the majority of today's machine learning as a service (MLaaS). Retraining the model on the original data, excluding the data to be forgotten (a.k.a forgetting data), is a common approach to class unlearning. However, the availability of original data during the unlearning phase is not always guaranteed, leading to the exploration of class unlearning with restricted data access. While current unlearning methods with restricted data access usually generate proxy sample via the trained neural network classifier, they typically focus on training and forgetting balanced data. However, the imbalanced original data can cause trouble for these proxies and unlearning, particularly when the forgetting data consists predominantly of the majority class. To address this issue, we propose the GENerative Imbalanced Unlearning (GENIU) framework. GENIU utilizes a Variational Autoencoder (VAE) to concurrently train a proxy generator alongside the original model. These generated proxies accurately represent each class and are leveraged in the unlearning phase, eliminating the reliance on the original training data. To further mitigate the performance degradation resulting from forgetting the majority class, we introduce an in-batch tuning strategy that works with the generated proxies. GENIU is the first practical framework for class unlearning in imbalanced data settings and restricted data access, ensuring the preservation of essential information for future unlearning. Experimental results confirm the superiority of GENIU over existing methods, establishing its effectiveness in empirical scenarios.