Research map / Label-efficient and robust learning / Self-supervised and contrastive
Learning a Self-Expressive Network for Subspace Clustering
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 |
|---|---|---|---|
| Use SENet instead of EnSC | subspace clustering · MNIST-full | ACC: 0.980 -> 0.968 | Hurt |
| Use SENet instead of NCSC | subspace clustering · Fashion-MNIST-full | ACC: 0.721 -> 0.697 | Hurt |
| Use SENet instead of ESC | subspace clustering · CIFAR-10 | ACC: 0.653 -> 0.765 | Helped |
| Use SENet instead of ESC | subspace clustering · EMNIST | ACC: 0.732 -> 0.721 | Hurt |
| Add a learnable soft-thresholding layer (ablation) | SENet · CIFAR-10 | ACC: 0.610 -> 0.765 | Helped |
| Add a learnable soft-thresholding layer (ablation) | SENet · CIFAR-10 | NMI: 0.588 -> 0.655 | Helped |
| Add a learnable soft-thresholding layer (ablation) | SENet · CIFAR-10 | ARI: 0.424 -> 0.573 | Helped |
| Increase network depth from 1 to 3 layers (ablation) | SENet · CIFAR-10 | ACC: 0.739 $\pm$ 0.002 -> 0.755 $\pm$ 0.016 | Helped |
| Increase network depth beyond 3 layers (ablation) | SENet · CIFAR-10 | ACC: 0.755 $\pm$ 0.016 -> 0.751 $\pm$ 0.013 | No clear effect |
| Increase network width (ablation) | SENet · CIFAR-10 | ACC | Helped |
| Use a batch size near 100 rather than an extreme batch size (ablation) | SENet training · CIFAR-10 | ACC | Helped |
Closest papers by meaning
- A Critique of Self-Expressive Deep Subspace Clustering (ICLR 2021 · similarity 0.92)
- Deep Subspace Clustering Networks (NeurIPS 2017 · similarity 0.92)
- Exploring a Principled Framework for Deep Subspace Clustering (ICLR 2025 · similarity 0.90)
- Partial Multi-View Clustering via Self-Supervised Network (AAAI 2024 · similarity 0.85)
- Customized Multiple Clustering via Multi-Modal Subspace Proxy Learning (NeurIPS 2024 · similarity 0.85)
- Clustering Properties of Self-Supervised Learning (ICML 2025 · similarity 0.85)
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.