Research map / Label-efficient and robust learning / Semi-supervised learning
Dual-Constrained Deep Semi-Supervised Coupled Factorization Network with Enriched Prior
Yan Zhang, Zhao Zhang, Yang Wang, Li Zhang, Li Zhang, Shuicheng Yan, Meng Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| DS2CF-Net instead of the best prior semi-supervised baseline RS2ACF | DS2CF-Net · COIL100 clustering | averaged AC: 0.8126 -> 0.8479 | Improved |
| DS2CF-Net instead of RS2ACF | DS2CF-Net · USPS clustering | averaged AC: 0.6780 -> 0.7276 | Improved |
| DS2CF-Net instead of RS2ACF | DS2CF-Net · AR clustering | averaged F-score: 0.5061 -> 0.5414 | Improved |
| add label prediction, dual constraints and fine-tuning over the unsupervised deep DSCF-Net | DS2CF-Net · Fashion MNIST clustering | averaged AC: 0.5784 -> 0.7026 | Improved |
| increase proportion of labeled samples (10% to 90%) (ablation) | semi-supervised MF methods · image databases, 3 categories | clustering accuracy | Improved |
| use semi-supervised DS2CF-Net rather than unsupervised DSCF-Net at 10% labels (ablation) | DS2CF-Net · AR, ETH80, COIL100, USPS | clustering accuracy | Hurt |
| increase number of layers (1 to 10) with the error-correcting coupled design (ablation) | DS2CF-Net · image databases, 3 categories | clustering accuracy | Improved |
| increase layers beyond 4 in plain stacked MNMF, MCF, GMCF, DSNMF (ablation) | MNMF/MCF/GMCF/DSNMF · image databases | clustering accuracy: passes 4 layers | Hurt |
| stack layers deeper (convergence per layer) (ablation) | DS2CF-Net · MIT CBCL, Fashion MNIST | iterations to converge: within 5 iterations in the 3rd layer | Improved |
| multilayer MNMF, MCF, GMCF versus single-layer models (ablation) | MNMF/MCF/GMCF · Fashion MNIST | clustering accuracy | Hurt |
Closest papers by meaning
- Partial-Label and Structure-constrained Deep Coupled Factorization Network (AAAI 2021 · similarity 0.97)
- A Deep Matrix Factorization Method for Learning Attribute Representations (TPAMI 2016 · similarity 0.86)
- Two-Stage Label Embedding via Neural Factorization Machine for Multi-Label Classification (AAAI 2019 · similarity 0.85)
- Constrained Factorization with Diagonal Scaling: Rank-Revealing Training and Pruning (NeurIPS 2026 · similarity 0.85)
- Two Sparse Matrices are Better than One: Sparsifying Neural Networks with Double Sparse Factorization (ICLR 2025 · similarity 0.84)
- Deep Weight Factorization: Sparse Learning Through the Lens of Artificial Symmetries (ICLR 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
