Efficient Machine Unlearning by Model Splitting and Core Sample Selection

TL;DR

This paper proposes MaxRR, a novel approach for efficient and verifiable machine unlearning that either achieves exact unlearning or approximates full retraining, addressing efficiency and verification challenges.

Contribution

Introduction of MaxRR, a generalized unlearning metric and training procedure enabling efficient, verifiable, and often exact unlearning in machine learning models.

Findings

MaxRR enables efficient unlearning with properties close to full retraining.

The approach provides a generalized metric for more precise unlearning evaluation.

MaxRR supports exact unlearning in many cases.

Abstract

Machine unlearning is essential for meeting legal obligations such as the right to be forgotten, which requires the removal of specific data from machine learning models upon request. While several approaches to unlearning have been proposed, existing solutions often struggle with efficiency and, more critically, with the verification of unlearning - particularly in the case of weak unlearning guarantees, where verification remains an open challenge. We introduce a generalized variant of the standard unlearning metric that enables more efficient and precise unlearning strategies. We also present an unlearning-aware training procedure that, in many cases, allows for exact unlearning. We term our approach MaxRR. When exact unlearning is not feasible, MaxRR still supports efficient unlearning with properties closely matching those achieved through full retraining.