Research map / Learning theory and algorithms / Clustering
A Critique of Self-Expressive Deep Subspace Clustering
Read the paperCheck 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 the self-expressive loss to autoencoder training (ablation) | SEDSC · Extended Yale-B (38 faces), without post-processing | clustering accuracy: 71.96% -> 59.09% | Hurt |
| Add the self-expressive loss to autoencoder training (ablation) | SEDSC · COIL100, without post-processing | clustering accuracy: 44.84% -> 45.67% | Helped |
| Add the self-expressive loss to autoencoder training (ablation) | SEDSC · ORL, without post-processing | clustering accuracy: 73.73% -> 73.50% | Hurt |
| Add post-processing to the coefficient matrix (ablation) | Full SEDSC · Extended Yale-B (38 faces) | clustering accuracy: 59.09% -> 96.79% | Helped |
| Add post-processing to the coefficient matrix (ablation) | Full SEDSC · COIL100 | clustering accuracy: 45.67% -> 64.96% | Helped |
| Add post-processing to the coefficient matrix (ablation) | Full SEDSC · ORL | clustering accuracy: 73.50% -> 84.10% | Helped |
| Add post-processing consistently across methods (ablation) | SEDSC and comparison methods · Extended Yale-B (38 faces), COIL100, and ORL | clustering accuracy | Mixed |
Closest papers by meaning
- Exploring a Principled Framework for Deep Subspace Clustering (ICLR 2025 · similarity 0.94)
- Deep Subspace Clustering Networks (NeurIPS 2017 · similarity 0.94)
- Efficient Deep Embedded Subspace Clustering (CVPR 2022 · similarity 0.93)
- Neural Collaborative Subspace Clustering (ICML 2019 · similarity 0.93)
- Learning a Self-Expressive Network for Subspace Clustering (CVPR 2021 · similarity 0.92)
- Customized Multiple Clustering via Multi-Modal Subspace Proxy Learning (NeurIPS 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.