Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
194 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

Refining Latent Homophilic Structures over Heterophilic Graphs for Robust Graph Convolution Networks (2312.16418v1)

Published 27 Dec 2023 in cs.LG, cs.AI, and cs.SI

Abstract: Graph convolution networks (GCNs) are extensively utilized in various graph tasks to mine knowledge from spatial data. Our study marks the pioneering attempt to quantitatively investigate the GCN robustness over omnipresent heterophilic graphs for node classification. We uncover that the predominant vulnerability is caused by the structural out-of-distribution (OOD) issue. This finding motivates us to present a novel method that aims to harden GCNs by automatically learning Latent Homophilic Structures over heterophilic graphs. We term such a methodology as LHS. To elaborate, our initial step involves learning a latent structure by employing a novel self-expressive technique based on multi-node interactions. Subsequently, the structure is refined using a pairwisely constrained dual-view contrastive learning approach. We iteratively perform the above procedure, enabling a GCN model to aggregate information in a homophilic way on heterophilic graphs. Armed with such an adaptable structure, we can properly mitigate the structural OOD threats over heterophilic graphs. Experiments on various benchmarks show the effectiveness of the proposed LHS approach for robust GCNs.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (51)
  1. Mixhop: Higher-order graph convolutional architectures via sparsified neighborhood mixing. In international conference on machine learning, 21–29. PMLR.
  2. Gephi: an open source software for exploring and manipulating networks. In Proceedings of the international AAAI conference on web and social media, volume 3.
  3. Evasion attacks against machine learning at test time. In Joint European conference on machine learning and knowledge discovery in databases. Springer.
  4. Beyond low-frequency information in graph convolutional networks. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 35, 3950–3957.
  5. The Frobenius norm and the commutator. Linear algebra and its applications, 429(8-9): 1864–1885.
  6. Embedding both finite and infinite communities on graphs. IEEE computational intelligence magazine, 14(3): 39–50.
  7. Adversarial attack on graph structured data. In International conference on machine learning, 1115–1124. PMLR.
  8. Finding structure with randomness: Probabilistic algorithms for constructing approximate matrix decompositions. SIAM review, 53(2): 217–288.
  9. Inductive Representation Learning on Large Graphs. In NIPS.
  10. Block modeling-guided graph convolutional neural networks. In Proceedings of the AAAI conference on artificial intelligence, volume 36, 4022–4029.
  11. Universal graph convolutional networks. Advances in Neural Information Processing Systems, 34: 10654–10664.
  12. Node similarity preserving graph convolutional networks. In Proceedings of the 14th ACM international conference on web search and data mining, 148–156.
  13. Graph structure learning for robust graph neural networks. In Proceedings of the 26th ACM SIGKDD international conference on knowledge discovery & data mining, 66–74.
  14. Variational graph auto-encoders. arXiv preprint arXiv:1611.07308.
  15. Semi-Supervised Classification with Graph Convolutional Networks. In International Conference on Learning Representations.
  16. Fast Markov clustering algorithm based on belief dynamics. IEEE Transactions on Cybernetics.
  17. Restructuring Graph for Higher Homophily via Adaptive Spectral Clustering. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 37, 8622–8630.
  18. Generative Causal Explanations for Graph Neural Networks. In Proceedings of the 38th International Conference on Machine Learning.
  19. Non-local graph neural networks. IEEE Transactions on Pattern Analysis and Machine Intelligence.
  20. Towards unsupervised deep graph structure learning. In Proceedings of the ACM Web Conference 2022, 1392–1403.
  21. Beyond smoothing: Unsupervised graph representation learning with edge heterophily discriminating. In Proceedings of the AAAI conference on artificial intelligence, volume 37, 4516–4524.
  22. Is Heterophily A Real Nightmare For Graph Neural Networks on Performing Node Classification?
  23. Geom-GCN: Geometric Graph Convolutional Networks. In International Conference on Learning Representations.
  24. 3D-IDS: Doubly Disentangled Dynamic Intrusion Detection. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 1965–1977.
  25. Fast Community Detection based on Graph Autoencoder Reconstruction. In 2022 7th International Conference on Big Data Analytics (ICBDA), 265–271. IEEE.
  26. VGAER: graph neural network reconstruction based community detection. arXiv preprint arXiv:2201.04066.
  27. Multi-scale attributed node embedding. Journal of Complex Networks, 9(2): cnab014.
  28. Breaking the Limit of Graph Neural Networks by Improving the Assortativity of Graphs with Local Mixing Patterns. In KDD.
  29. Adversarial graph augmentation to improve graph contrastive learning. Advances in Neural Information Processing Systems, 34: 15920–15933.
  30. Social influence analysis in large-scale networks. In Proceedings of the 15th ACM SIGKDD international conference on Knowledge discovery and data mining, 807–816.
  31. Deep graph infomax. ICLR (Poster), 2(3): 4.
  32. Graph Attention Networks. In International Conference on Learning Representations.
  33. Powerful Graph Convolutional Networks with Adaptive Propagation Mechanism for Homophily and Heterophily. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 36, 4210–4218.
  34. Tree decomposed graph neural network. In Proceedings of the 30th ACM International Conference on Information & Knowledge Management.
  35. ACMP: Allen-cahn message passing with attractive and repulsive forces for graph neural networks. In The Eleventh International Conference on Learning Representations.
  36. How ground-truth label helps link prediction in heterogeneous graphs. In Proceedings of the 2023 2nd International Conference on Algorithms, Data Mining, and Information Technology, 6–12.
  37. Node Classification Beyond Homophily: Towards a General Solution. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2862–2873.
  38. Two sides of the same coin: Heterophily and oversmoothing in graph convolutional neural networks. arXiv preprint arXiv:2102.06462.
  39. Diverse Message Passing for Attribute with Heterophily. In Beygelzimer, A.; Dauphin, Y.; Liang, P.; and Vaughan, J. W., eds., Advances in Neural Information Processing Systems.
  40. Graph pointer neural networks. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 36, 8832–8839.
  41. Revisiting semi-supervised learning with graph embeddings. In International conference on machine learning, 40–48. PMLR.
  42. Node2Seq: Towards Trainable Convolutions in Graph Neural Networks. CoRR.
  43. Adversarial Feature Selection Against Evasion Attacks. IEEE Transactions on Cybernetics, 46.
  44. Self-supervised structure learning for crack detection based on cycle-consistent generative adversarial networks. Journal of Computing in Civil Engineering, 34(3): 04020004.
  45. Gnnguard: Defending graph neural networks against adversarial attacks. Advances in neural information processing systems, 33: 9263–9275.
  46. Graph neural networks for graphs with heterophily: A survey. arXiv preprint arXiv:2202.07082.
  47. Finding the Missing-half: Graph Complementary Learning for Homophily-prone and Heterophily-prone Graphs. arXiv preprint arXiv:2306.07608.
  48. Robust graph convolutional networks against adversarial attacks. In Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery & data mining, 1399–1407.
  49. How does heterophily impact the robustness of graph neural networks? theoretical connections and practical implications. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2637–2647.
  50. Beyond homophily in graph neural networks: Current limitations and effective designs. Advances in Neural Information Processing Systems, 33: 7793–7804.
  51. Deep Graph Contrastive Representation Learning. CoRR.
Citations (3)
View on Semantic Scholar

Summary

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

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now