Graph Unlearning with Efficient Partial Retraining
This addresses the need for efficient data removal in GNNs for applications where training data may be unreliable, though it appears incremental as it builds on retraining-based unlearning methods.
The paper tackles the problem of graph neural networks (GNNs) being trained on undesirable data, which degrades performance, by proposing GraphRevoker, a graph unlearning framework that improves model utility through graph property-aware sharding and contrastive sub-model aggregation, achieving superior results in experiments.
Graph Neural Networks (GNNs) have achieved remarkable success in various real-world applications. However, GNNs may be trained on undesirable graph data, which can degrade their performance and reliability. To enable trained GNNs to efficiently unlearn unwanted data, a desirable solution is retraining-based graph unlearning, which partitions the training graph into subgraphs and trains sub-models on them, allowing fast unlearning through partial retraining. However, the graph partition process causes information loss in the training graph, resulting in the low model utility of sub-GNN models. In this paper, we propose GraphRevoker, a novel graph unlearning framework that better maintains the model utility of unlearnable GNNs. Specifically, we preserve the graph property with graph property-aware sharding and effectively aggregate the sub-GNN models for prediction with graph contrastive sub-model aggregation. We conduct extensive experiments to demonstrate the superiority of our proposed approach.