Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
97 tokens/sec
GPT-4o
53 tokens/sec
Gemini 2.5 Pro Pro
44 tokens/sec
o3 Pro
5 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Investigating and Mitigating the Side Effects of Noisy Views for Self-Supervised Clustering Algorithms in Practical Multi-View Scenarios (2303.17245v4)

Published 30 Mar 2023 in cs.LG and cs.CV

Abstract: Multi-view clustering (MVC) aims at exploring category structures among multi-view data in self-supervised manners. Multiple views provide more information than single views and thus existing MVC methods can achieve satisfactory performance. However, their performance might seriously degenerate when the views are noisy in practical multi-view scenarios. In this paper, we formally investigate the drawback of noisy views and then propose a theoretically grounded deep MVC method (namely MVCAN) to address this issue. Specifically, we propose a novel MVC objective that enables un-shared parameters and inconsistent clustering predictions across multiple views to reduce the side effects of noisy views. Furthermore, a two-level multi-view iterative optimization is designed to generate robust learning targets for refining individual views' representation learning. Theoretical analysis reveals that MVCAN works by achieving the multi-view consistency, complementarity, and noise robustness. Finally, experiments on extensive public datasets demonstrate that MVCAN outperforms state-of-the-art methods and is robust against the existence of noisy views.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (73)
  1. Deep multimodal subspace clustering networks. IEEE Journal of Selected Topics in Signal Processing, 12(6):1601–1614, 2018.
  2. Pierre Baldi. Autoencoders, unsupervised learning, and deep architectures. In Proceedings of ICML Workshop on Unsupervised and Transfer Learning, pages 37–49, 2012.
  3. Joint stage recognition and anatomical annotation of drosophila gene expression patterns. Bioinformatics, 28(12):i16–i24, 2012.
  4. Diversity-induced multi-view subspace clustering. In CVPR, pages 586–594, 2015.
  5. Deep multiview clustering by contrasting cluster assignments. In ICCV, pages 16752–16761, 2023.
  6. Cross-view topology based consistent and complementary information for deep multi-view clustering. In ICCV, pages 19440–19451, 2023.
  7. One2multi graph autoencoder for multi-view graph clustering. In WWW, pages 3070–3076, 2020.
  8. Learning generative visual models from few training examples: An incremental bayesian approach tested on 101 object categories. In CVPR, pages 178–178, 2004.
  9. Deep sparse rectifier neural networks. In AISTATS, pages 315–323, 2011.
  10. Generative adversarial nets. In NeurIPS, pages 2672–2680, 2014.
  11. Improved deep embedded clustering with local structure preservation. In IJCAI, pages 1753–1759, 2017.
  12. Comment-based multi-view clustering of web 2.0 items. In WWW, pages 771–782, 2014.
  13. Partially view-aligned clustering. In NeurIPS, pages 2892–2902, 2020.
  14. Self-supervised graph attention networks for deep weighted multi-view clustering. In AAAI, pages 7936–7943, 2023.
  15. Deep incomplete multi-view clustering with cross-view partial sample and prototype alignment. In CVPR, pages 11600–11609, 2023.
  16. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014.
  17. Reciprocal multi-layer subspace learning for multi-view clustering. In ICCV, pages 8172–8180, 2019a.
  18. scbridge embraces cell heterogeneity in single-cell rna-seq and atac-seq data integration. Nature Communications, 14(1):6045, 2023.
  19. Deep adversarial multi-view clustering network. In IJCAI, pages 2952–2958, 2019b.
  20. COMPLETER: Incomplete multi-view clustering via contrastive prediction. In CVPR, pages 11174–11183, 2021.
  21. Human action recognition and retrieval using sole depth information. In ACM MM, pages 1053–1056, 2012.
  22. Contrastive multi-view kernel learning. TPAMI, pages 1–15, 2023.
  23. Late fusion incomplete multi-view clustering. TPAMI, 41(10):2410–2423, 2018.
  24. One pass late fusion multi-view clustering. In ICML, pages 6850–6859, 2021.
  25. Decoupled contrastive multi-view clustering with high-order random walks. arXiv preprint arXiv:2308.11164, 2023.
  26. Laurens van der Maaten and Geoffrey Hinton. Visualizing data using t𝑡titalic_t-SNE. JMLR, 9:2579–2605, 2008.
  27. James MacQueen. Some methods for classification and analysis of multivariate observations. In Proceedings of the Berkeley Symposium on Mathematical Statistics and Probability, pages 281–297, 1967.
  28. On using nearly-independent feature families for high precision and confidence. Machine Learning, 92:457–477, 2013.
  29. Columbia object image library (coil-100). http://www1. cs. columbia. edu/CAVE/software/softlib/coil-100. php, 1996.
  30. On spectral clustering: Analysis and an algorithm. In NeurIPS, pages 849–856, 2001.
  31. Parameter-free auto-weighted multiple graph learning: a framework for multiview clustering and semi-supervised classification. In IJCAI, pages 1881–1887, 2016.
  32. Multiview clustering via adaptively weighted procrustes. In KDD, pages 2022–2030, 2018.
  33. Analyzing the effectiveness and applicability of co-training. In CIKM, pages 86–93, 2000.
  34. Representation learning with contrastive predictive coding. arXiv preprint arXiv:1807.03748, 2018.
  35. PyTorch: An imperative style, high-performance deep learning library. In NeurIPS, pages 8024–8035, 2019.
  36. COMIC: Multi-view clustering without parameter selection. In ICML, pages 5092–5101, 2019.
  37. Xai beyond classification: Interpretable neural clustering. JMLR, 23(1):227–254, 2022.
  38. Multi-omic and multi-view clustering algorithms: review and cancer benchmark. Nucleic acids research, 46(20):10546–10562, 2018.
  39. Deep clustering: A comprehensive survey. arXiv preprint arXiv:2210.04142, 2022.
  40. Adapting visual category models to new domains. In ECCV, pages 213–226, 2010.
  41. Unified one-step multi-view spectral clustering. TKDE, 35(6):6449–6460, 2022.
  42. Deep safe multi-view clustering: Reducing the risk of clustering performance degradation caused by view increase. In CVPR, pages 202–211, 2022a.
  43. Deep safe incomplete multi-view clustering: Theorem and algorithm. In ICML, pages 21090–21110, 2022b.
  44. Contrastive multiview coding. In ECCV, pages 776–794, 2020.
  45. Reconsidering representation alignment for multi-view clustering. In CVPR, pages 1255–1265, 2021.
  46. Kernel-based weighted multi-view clustering. In ICDM, pages 675–684, 2012.
  47. Multigraph spectral clustering for joint content delivery and scheduling in beam-free satellite communications. In ICASSP, pages 8802–8806, 2020.
  48. Multi-view clustering and feature learning via structured sparsity. In ICML, pages 352–360, 2013.
  49. Generative partial multi-view clustering with adaptive fusion and cycle consistency. TIP, 30:1771–1783, 2021.
  50. Multi-view clustering via late fusion alignment maximization. In IJCAI, pages 3778–3784, 2019.
  51. Highly-efficient incomplete large-scale multi-view clustering with consensus bipartite graph. In CVPR, pages 9776–9785, 2022.
  52. Iterative views agreement: an iterative low-rank based structured optimization method to multi-view spectral clustering. In IJCAI, pages 2153–2159, 2016.
  53. Multiview spectral clustering via structured low-rank matrix factorization. TNNLS, 29(10):4833–4843, 2018.
  54. Incomplete multiview spectral clustering with adaptive graph learning. TCYB, 50(4):1418–1429, 2018.
  55. DIMC-net: Deep incomplete multi-view clustering network. In ACM MM, pages 3753–3761, 2020.
  56. Highly confident local structure based consensus graph learning for incomplete multi-view clustering. In CVPR, pages 15712–15721, 2023.
  57. Unsupervised deep embedding for clustering analysis. In ICML, pages 478–487, 2016.
  58. Joint deep multi-view learning for image clustering. TKDE, 33(11):3594–3606, 2020.
  59. Adversarial incomplete multi-view clustering. In IJCAI, pages 3933–3939, 2019.
  60. Deep incomplete multi-view clustering via mining cluster complementarity. In AAAI, pages 8761–8769, 2022.
  61. UNTIE: Clustering analysis with disentanglement in multi-view information fusion. Information Fusion, 100:101937, 2023a.
  62. Self-supervised discriminative feature learning for deep multi-view clustering. TKDE, 35(7):7470–7482, 2023b.
  63. GCFAgg: Global and cross-view feature aggregation for multi-view clustering. In CVPR, pages 19863–19872, 2023.
  64. Robust multi-view clustering with incomplete information. TPAMI, 45(1):1055–1069, 2022.
  65. Uniform distribution non-negative matrix factorization for multiview clustering. TCYB, pages 3249–3262, 2021.
  66. Multi-view clustering with noisy views. In Proceedings of the International Conference on Computer Science and Artificial Intelligence, pages 339–344, 2018.
  67. Multiview consensus graph clustering. TIP, 28(3):1261–1270, 2018.
  68. Graph structure fusion for multiview clustering. TKDE, 31(10):1984–1993, 2019.
  69. Latent multi-view subspace clustering. In CVPR, pages 4279–4287, 2017.
  70. Generalized latent multi-view subspace clustering. TPAMI, 42(1):86–99, 2018.
  71. Deep partial multi-view learning. TPAMI, 44(5):2402–2415, 2020.
  72. End-to-end adversarial-attention network for multi-modal clustering. In CVPR, pages 14619–14628, 2020.
  73. Dual shared-specific multiview subspace clustering. TCYB, 50(8):3517–3530, 2019.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (7)
  1. Jie Xu (467 papers)
  2. Yazhou Ren (35 papers)
  3. Xiaolong Wang (243 papers)
  4. Lei Feng (190 papers)
  5. Zheng Zhang (488 papers)
  6. Gang Niu (125 papers)
  7. Xiaofeng Zhu (56 papers)
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