MUC: Machine Unlearning for Contrastive Learning with Black-box Evaluation

TL;DR

This paper introduces MUC, a framework for machine unlearning in contrastive learning models, featuring a novel alignment calibration method that improves unlearning effectiveness and provides easy black-box evaluation.

Contribution

It proposes Alignment Calibration, a new method tailored for contrastive learning unlearning, and develops evaluation tools for verifying unlearning effects.

Findings

AC achieves state-of-the-art unlearning performance

AC enables clear visualization of unlearning effects

Method works effectively on SimCLR, MoCo, and CLIP

Abstract

Machine unlearning offers effective solutions for revoking the influence of specific training data on pre-trained model parameters. While existing approaches address unlearning for classification and generative models, they overlook an important category of machine learning models: contrastive learning (CL) methods. This paper addresses this gap by introducing the Machine Unlearning for Contrastive Learning (MUC) framework and adapting existing methods. We identify limitations in current approaches, noting that several methods perform inadequately as unlearners and that existing evaluation tools insufficiently validate unlearning effects in contrastive learning. To address these issues, we propose Alignment Calibration (AC), a novel method that explicitly considers contrastive learning properties and optimizes towards new auditing metrics for easy verification of unlearning. Through empirical comparisons with baseline methods on SimCLR, MoCo, and CLIP, we demonstrate that AC: (1) achieves state-of-the-art performance, approximating exact unlearning (retraining); (2) enables data owners to clearly visualize unlearning effects through black-box evaluation. The code is available at https://github.com/EhanW/Alignment-Calibration.