Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
158 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Graph Unlearning with Efficient Partial Retraining (2403.07353v2)

Published 12 Mar 2024 in cs.LG and cs.CR

Abstract: 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.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (35)
  1. Streaming graph partitioning: an experimental study. In VLDB.
  2. Machine unlearning. In IEEE S&P.
  3. Jonathan Brophy and Daniel Lowd. 2021. Machine unlearning for random forests. In ICML.
  4. Yinzhi Cao and Junfeng Yang. 2015. Towards making systems forget with machine unlearning. In IEEE S&P.
  5. Recommendation unlearning. In WWW.
  6. Knowledge-enhanced black-box attacks for recommendations. In KDD.
  7. Graph unlearning. In ACM CCS.
  8. GNNDelete: A General Unlearning Strategy for Graph Neural Networks. In ICLR.
  9. Graph trend filtering networks for recommendation. In SIGIR.
  10. Untargeted Black-box Attacks for Social Recommendations. arXiv preprint (2023).
  11. Jointly attacking graph neural network and its explanations. In ICDE.
  12. Predict then Propagate: Graph Neural Networks meet Personalized PageRank. In ICLR.
  13. Eternal sunshine of the spotless net: Selective forgetting in deep networks. In CVPR.
  14. Certified data removal from machine learning models. In ICML.
  15. Inductive representation learning on large graphs. In NeurIPS.
  16. George Karypis and Vipin Kumar. 1997. METIS: A software package for partitioning unstructured graphs, partitioning meshes, and computing fill-reducing orderings of sparse matrices. Technical Report (1997).
  17. Thomas N Kipf and Max Welling. 2016. Semi-Supervised Classification with Graph Convolutional Networks. In ICLR.
  18. Wenqing Lin. 2021. Large-scale network embedding in apache spark. In KDD.
  19. Graph adversarial attack via rewiring. In KDD.
  20. Benedek Rozemberczki and Rik Sarkar. 2020. Characteristic functions on graphs: Birds of a feather, from statistical descriptors to parametric models. In CIKM.
  21. Isabelle Stanton and Gabriel Kliot. 2012. Streaming graph partitioning for large distributed graphs. In KDD.
  22. On the necessity of auditable algorithmic definitions for machine unlearning. In USENIX Security.
  23. Graph Attention Networks. In ICLR.
  24. Fast graph condensation with structure-based neural tangent kernel. In WWW.
  25. Machine unlearning of features and labels. In NDSS.
  26. Disentangled contrastive learning for social recommendation. In CIKM.
  27. GIF: A General Graph Unlearning Strategy via Influence Function. In WWW.
  28. Distributed power-law graph computing: Theoretical and empirical analysis. NeurIPS.
  29. Representation learning on graphs with jumping knowledge networks. In ICML.
  30. ARCANE: An Efficient Architecture for Exact Machine Unlearning.. In IJCAI.
  31. Revisiting semi-supervised learning with graph embeddings. In ICML.
  32. GraphSAINT: Graph Sampling Based Inductive Learning Method. In ICLR.
  33. Graph edge partitioning via neighborhood heuristic. In KDD.
  34. Linear-Time Graph Neural Networks for Scalable Recommendations. In WWW.
  35. Graph contrastive learning with adaptive augmentation. In WWW.
Citations (3)

Summary

We haven't generated a summary for this paper yet.