Research map / Label-efficient and robust learning / Self-supervised and contrastive
Self-Supervised Convolutional Subspace Clustering Network
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| add dual self-supervision losses L3 and L4 to the DSC-style objective (ablation) | S²ConvSCN-ℓ1 · Extended Yale B | clustering error (%): 2.23 -> 1.18 (10 subjects, mean); 3.33 -> 1.52 (38 subjects, mean) | Helped |
| add spectral clustering loss L3 only (ablation) | S²ConvSCN-ℓ1 · Extended Yale B | clustering error (%): 2.23 -> 1.58 (10 subjects, mean); 3.33 -> 2.75 (38 subjects, mean) | Helped |
| add classification loss L4 only (ablation) | S²ConvSCN-ℓ1 · Extended Yale B | clustering error (%): 2.23 -> 1.32 (10 subjects, mean); 3.33 -> 1.92 (38 subjects, mean) | Helped |
| combine classification loss L4 with spectral clustering loss L3 rather than using L4 alone (ablation) | S²ConvSCN-ℓ1 · Extended Yale B | clustering error (%): 1.32 -> 1.18 (10 subjects, mean); 1.92 -> 1.52 (38 subjects, mean) | Helped |
| use S²ConvSCN-ℓ1 instead of DASC | subspace clustering · Extended Yale B, 38 subjects | clustering error (%): 1.44 -> 1.52 | Hurt |
| use S²ConvSCN-ℓ1 instead of DASC | subspace clustering · ORL | clustering error (%): 11.75 -> 10.50 | Helped |
| use S²ConvSCN-ℓ1 instead of DASC | subspace clustering · COIL20 | clustering error (%): 3.61 -> 2.14 | Helped |
| use S²ConvSCN-ℓ1 instead of DSC-ℓ2 | subspace clustering · COIL100 | clustering error (%): 30.96 -> 26.67 | Helped |
Closest papers by meaning
- Learning a Self-Expressive Network for Subspace Clustering (CVPR 2021 · similarity 0.91)
- Deep Subspace Clustering Networks (NeurIPS 2017 · similarity 0.90)
- LRSC: Learning Representations for Subspace Clustering (AAAI 2021 · similarity 0.90)
- A Critique of Self-Expressive Deep Subspace Clustering (ICLR 2021 · similarity 0.90)
- Exploring a Principled Framework for Deep Subspace Clustering (ICLR 2025 · similarity 0.89)
- Partial Multi-View Clustering via Self-Supervised Network (AAAI 2024 · similarity 0.87)
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.