Efficient Clustering Network Based on Matrix Factorization
Source abstract
Abstract
Contrastive learning is a significant research direction in the field of deep learning. However, existing data augmentation methods often lead to issues such as semantic drift in generated views while the complexity of model pre-training limits further improvement in the performance of existing methods. To address these challenges, we propose the Efficient Clustering Network based on Matrix Factorization (ECN-MF). Specifically, we design a batched low-rank Singular Value Decomposition (SVD) algorithm for data augmentation to eliminate redundant information and uncover major patterns of variation and key information in the data. Additionally, we design a Mutual Information-Enhanced Clustering Module (MI-ECM) to accelerate the training process by leveraging a simple architecture to bring samples from the same cluster closer while pushing samples from other clusters apart. Extensive experiments on six datasets demonstrate that ECN-MF exhibits more effective performance compared to state-of-the-art algorithms.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Advanced Computing and Algorithms · Face and Expression Recognition · Advanced Clustering Algorithms Research
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (42)
Show source-provided references
1. WSNMF: Weighted symmetric nonnegative matrix factorization for attributed graph clustering10.1016/j.neucom.2023.127041 ↗
2. A deep semi-supervised community detection based on point-wise mutual information
3. Multi-view contrastive graph clustering
4. Attributed graph clustering: A deep attentional embedding approach
5. Structural deep clustering network
6. Deep fusion clustering network
7. Attention-driven graph clustering network
8. Symmetric graph convolutional autoencoder for unsupervised graph representation learning
9. Multi-view attribute graph convolution networks for clustering
10. Learning graph embedding with adversarial training methods10.1109/TCYB.2019.2932096 ↗
11. Adversarial graph embedding for ensemble clustering
12. Adaptive graph encoder for attributed graph embedding
13. Contrastive multi-view representation learning on graphs
14. Multi-view graph embedding clustering network: Joint self-supervision and block diagonal representation10.1016/j.neunet.2021.10.006 ↗
15. Deep graph clustering via dual correlation reduction
16. Multilayer graph contrastive clustering network10.1016/j.ins.2022.09.042 ↗
17. Multi-view clustering: A survey, big data mining and analytics10.26599/BDMA.2018.9020003 ↗
18. LightGCL: Simple yet effective graph contrastive learning for recommendation
19. Multi-view data clustering via non-negative matrix factorization with manifold regularization10.1007/s13042-021-01307-7 ↗
20. A simple framework for contrastive learning of visual representations
21. Barlow twins: Self-supervised learning via redundancy reduction
22. Graph contrastive clustering
23. Interpolation-based contrastive learning for few-label semi-supervised learning10.1109/TNNLS.2022.3186512 ↗
24. Graph contrastive learning with augmentations
25. SimGRACE: A simple framework for graph contrastive learning without data augmentation
26. Graph barlow twins: A self-supervised representation learning framework for graphs10.1016/j.knosys.2022.109631 ↗
27. Augmentation-free self-supervised learning on graphs
28. Attributed graph clustering with dual redundancy reduction
29. Rectified linear units improve restricted boltzmann machines
30. Rethinking graph auto-encoder models for attributed graph clustering
31. Multiple kernel clustering with neighbor-kernel subspace segmentation10.1109/TNNLS.2019.2919900 ↗
32. Fast parameter-free multi-view subspace clustering with consensus anchor guidance10.1109/TIP.2021.3131941 ↗
33. Late fusion multiple kernel clustering with proxy graph refinement10.1109/TNNLS.2021.3117403 ↗
34. GraphMAE2: A decoding-enhanced masked self-supervised graph learner
35. Towards K-means-friendly spaces: Simultaneous deep learning and clustering
36. Unsupervised deep embedding for clustering analysis
37. Simple spectral graph convolution
38. Scalable attributed-graph subspace clustering
39. Graph debiased contrastive learning with joint representation clustering
40. ProGCL: Rethinking hard negative mining in graph contrastive learning
41. Automated self-supervised learning for graphs
42. Self-contrastive graph diffusion network