Machine Unlearning Method Based On Projection Residual

TL;DR

This paper introduces a projection residual-based machine unlearning method using Newton iteration, enabling neural networks and linear models to forget specific data efficiently, complying with data privacy laws.

Contribution

It proposes a novel unlearning technique that is independent of training set size and achieves near-retraining thoroughness with linear computational cost.

Findings

Method effectively deletes data influence close to retraining

Computational cost is linear in feature dimension

Outperforms existing unlearning approaches in thoroughness

Abstract

Machine learning models (mainly neural networks) are used more and more in real life. Users feed their data to the model for training. But these processes are often one-way. Once trained, the model remembers the data. Even when data is removed from the dataset, the effects of these data persist in the model. With more and more laws and regulations around the world protecting data privacy, it becomes even more important to make models forget this data completely through machine unlearning. This paper adopts the projection residual method based on Newton iteration method. The main purpose is to implement machine unlearning tasks in the context of linear regression models and neural network models. This method mainly uses the iterative weighting method to completely forget the data and its corresponding influence, and its computational cost is linear in the feature dimension of the data. This method can improve the current machine learning method. At the same time, it is independent of the size of the training set. Results were evaluated by feature injection testing (FIT). Experiments show that this method is more thorough in deleting data, which is close to model retraining.