Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
143 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 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

ROG$_{PL}$: Robust Open-Set Graph Learning via Region-Based Prototype Learning (2402.18495v2)

Published 28 Feb 2024 in cs.LG

Abstract: Open-set graph learning is a practical task that aims to classify the known class nodes and to identify unknown class samples as unknowns. Conventional node classification methods usually perform unsatisfactorily in open-set scenarios due to the complex data they encounter, such as out-of-distribution (OOD) data and in-distribution (IND) noise. OOD data are samples that do not belong to any known classes. They are outliers if they occur in training (OOD noise), and open-set samples if they occur in testing. IND noise are training samples which are assigned incorrect labels. The existence of IND noise and OOD noise is prevalent, which usually cause the ambiguity problem, including the intra-class variety problem and the inter-class confusion problem. Thus, to explore robust open-set learning methods is necessary and difficult, and it becomes even more difficult for non-IID graph data.To this end, we propose a unified framework named ROG${PL}$ to achieve robust open-set learning on complex noisy graph data, by introducing prototype learning. In specific, ROG${PL}$ consists of two modules, i.e., denoising via label propagation and open-set prototype learning via regions. The first module corrects noisy labels through similarity-based label propagation and removes low-confidence samples, to solve the intra-class variety problem caused by noise. The second module learns open-set prototypes for each known class via non-overlapped regions and remains both interior and border prototypes to remedy the inter-class confusion problem.The two modules are iteratively updated under the constraints of classification loss and prototype diversity loss. To the best of our knowledge, the proposed ROG$_{PL}$ is the first robust open-set node classification method for graph data with complex noise.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (42)
  1. Temperature check: theory and practice for training models with softmax-cross-entropy losses. arXiv:2010.07344.
  2. A closer look at memorization in deep networks. In International conference on machine learning, 233–242. PMLR.
  3. Towards open set deep networks. In IEEE conference on computer vision and pattern recognition, 1563–1572.
  4. Fast, exact and multi-scale inference for semantic image segmentation with deep gaussian crfs. In Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016, Proceedings, Part VII 14, 402–418. Springer.
  5. Utilizing Molecular Network Information via Graph Convolutional Neural Networks to Predict Metastatic Event in Breast Cancer. In GMDS, 181–186.
  6. Active learning for crowdsourcing using knowledge transfer. In Proceedings of the AAAI Conference on Artificial Intelligence, 1809–1815.
  7. Neural message passing for quantum chemistry. In International conference on machine learning, 1263–1272. PMLR.
  8. Semi-supervised learning by entropy minimization. Advances in neural information processing systems, 17.
  9. Orthogonal graph neural networks. In Proceedings of the AAAI Conference on Artificial Intelligence, 3996–4004.
  10. Inductive representation learning on large graphs. Advances in neural information processing systems, 30.
  11. A baseline for detecting misclassified and out-of-distribution examples in neural networks. International Conference on Learning Representations.
  12. Open graph benchmark: Datasets for machine learning on graphs. Advances in neural information processing systems, 33: 22118–22133.
  13. End-to-end open-set semi-supervised node classification with out-of-distribution detection. In IJCAI, 2087–2093.
  14. Mining on manifolds: Metric learning without labels. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 7642–7651.
  15. Label propagation for deep semi-supervised learning. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 5070–5079.
  16. Efficient diffusion on region manifolds: Recovering small objects with compact cnn representations. In Proceedings of the IEEE conference on computer vision and pattern recognition, 2077–2086.
  17. Semi-supervised classification with graph convolutional networks. arXiv:1609.02907.
  18. Selective-supervised contrastive learning with noisy labels. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 316–325.
  19. Webvision database: Visual learning and understanding from web data. arXiv:1708.02862.
  20. Learning from noisy labels with distillation. In Proceedings of the IEEE international conference on computer vision, 1910–1918.
  21. Graph self-supervised learning: A survey. IEEE Transactions on Knowledge and Data Engineering, 35(6): 5879–5900.
  22. ProtoInfoMax: Prototypical Networks with Mutual Information Maximization for Out-of-Domain Detection. In Moens, M.-F.; Huang, X.; Specia, L.; and Yih, S. W.-t., eds., Findings of the Association for Computational Linguistics: EMNLP 2021, 1606–1617. Punta Cana, Dominican Republic: Association for Computational Linguistics.
  23. Pnp: Robust learning from noisy labels by probabilistic noise prediction. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 5311–5320.
  24. Ethereum fraud behavior detection based on graph neural networks. Computing, 1–28.
  25. Graph auto-encoder via neighborhood wasserstein reconstruction. arXiv:2202.09025.
  26. Prototype-based intent perception. IEEE Transactions on Multimedia.
  27. Standing up or standing by: Understanding bystanders’ proactive reporting responses to social media harassment. Information Systems Research, 32(2): 561–581.
  28. Openwgl: Open-world graph learning. In IEEE international conference on data mining, 681–690.
  29. NGC: A unified framework for learning with open-world noisy data. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 62–71.
  30. Learning from massive noisy labeled data for image classification. In Proceedings of the IEEE conference on computer vision and pattern recognition, 2691–2699.
  31. Robust classification with convolutional prototype learning. In Proceedings of the IEEE conference on computer vision and pattern recognition, 3474–3482.
  32. Revisiting semi-supervised learning with graph embeddings. In International conference on machine learning, 40–48. PMLR.
  33. Learning with biased complementary labels. In Proceedings of the European conference on computer vision (ECCV), 68–83.
  34. Open-category classification by adversarial sample generation. In International Joint Conference on Artificial Intelligence, 3357–3363.
  35. Understanding deep learning (still) requires rethinking generalization. Communications of the ACM, 64(3): 107–115.
  36. Denoising Multi-Similarity Formulation: A Self-Paced Curriculum-Driven Approach for Robust Metric Learning. In Proceedings of the AAAI Conference on Artificial Intelligence, 11183–11191.
  37. A Survey for Efficient Open Domain Question Answering. In Rogers, A.; Boyd-Graber, J.; and Okazaki, N., eds., Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 14447–14465. Toronto, Canada: Association for Computational Linguistics.
  38. A Dynamic Variational Framework for Open-World Node Classification in Structured Sequences. In 2022 IEEE International Conference on Data Mining (ICDM), 703–712. IEEE.
  39. G2Pxy: generative open-set node classification on graphs with proxy unknowns. In Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence, 4576–4583.
  40. Gman: A graph multi-attention network for traffic prediction. In AAAI Conference on Artificial Intelligence, 1234–1241.
  41. Robust graph representation learning for local corruption recovery. In Proceedings of the ACM Web Conference 2023, 438–448.
  42. Shift-Robust Node Classification via Graph Adversarial Clustering. arXiv:2203.15802.
Citations (1)
View on Semantic Scholar

Summary

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

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