Efficiently Forgetting What You Have Learned in Graph Representation Learning via Projection

TL;DR

This paper introduces PROJECTOR, a novel method for unlearning specific nodes in graph neural networks by projecting model weights onto a subspace, effectively removing node information while maintaining model performance.

Contribution

It presents a new projection-based approach for unlearning in GNNs that handles node dependency and guarantees complete removal of node information.

Findings

PROJECTOR effectively unlearns nodes in real-world datasets.

The method guarantees perfect data removal of unlearned nodes.

Empirical results show high efficiency and effectiveness of the approach.

Abstract

As privacy protection receives much attention, unlearning the effect of a specific node from a pre-trained graph learning model has become equally important. However, due to the node dependency in the graph-structured data, representation unlearning in Graph Neural Networks (GNNs) is challenging and less well explored. In this paper, we fill in this gap by first studying the unlearning problem in linear-GNNs, and then introducing its extension to non-linear structures. Given a set of nodes to unlearn, we propose PROJECTOR that unlearns by projecting the weight parameters of the pre-trained model onto a subspace that is irrelevant to features of the nodes to be forgotten. PROJECTOR could overcome the challenges caused by node dependency and enjoys a perfect data removal, i.e., the unlearned model parameters do not contain any information about the unlearned node features which is guaranteed by algorithmic construction. Empirical results on real-world datasets illustrate the effectiveness and efficiency of PROJECTOR.