Research map / Learning theory and algorithms / Matrix and tensor methods
Implicit Bias of Projected Subgradient Method Gives Provable Robust Recovery of Subspaces of Unknown Codimension
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 |
|---|---|---|---|
| Use DPCP-PSGM with unknown codimension instead of RSGM with unknown codimension | Robust subspace recovery on Washington DC Mall AVIRIS HSI · Washington DC Mall AVIRIS HSI, r = 80% | F1 score: 0 -> 0.994 | Helped |
| Use DPCP-PSGM with unknown codimension instead of DPCP-IRLS with unknown codimension | Robust subspace recovery on Washington DC Mall AVIRIS HSI · Washington DC Mall AVIRIS HSI, r = 80% | F1 score: 0 -> 0.994 | Helped |
| Use DPCP-PSGM with unknown codimension instead of RSGM with unknown codimension | Robust subspace recovery on Washington DC Mall AVIRIS HSI · Washington DC Mall AVIRIS HSI, r = 90% | F1 score: 0 -> 0.993 | Helped |
| Use DPCP-PSGM with unknown codimension instead of DPCP-IRLS with unknown codimension | Robust subspace recovery on Washington DC Mall AVIRIS HSI · Washington DC Mall AVIRIS HSI, r = 90% | F1 score: 0 -> 0.993 | Helped |
| Use an overestimated codimension with DPCP-PSGM rather than supplying the true codimension to RSGM | Robust subspace recovery on Washington DC Mall AVIRIS HSI · Washington DC Mall AVIRIS HSI, r = 80% | F1 score: 0.999 -> 0.994 | Hurt |
| Use an overestimated codimension with DPCP-PSGM rather than supplying the true codimension to DPCP-IRLS | Robust subspace recovery on Washington DC Mall AVIRIS HSI · Washington DC Mall AVIRIS HSI, r = 80% | F1 score: 1 -> 0.994 | Hurt |
| Use an overestimated codimension with DPCP-PSGM rather than supplying the true codimension to RSGM | Robust subspace recovery on Washington DC Mall AVIRIS HSI · Washington DC Mall AVIRIS HSI, r = 90% | F1 score: 0.993 -> 0.993 | No clear effect |
| Use an overestimated codimension with DPCP-PSGM rather than supplying the true codimension to DPCP-IRLS | Robust subspace recovery on Washington DC Mall AVIRIS HSI · Washington DC Mall AVIRIS HSI, r = 90% | F1 score: 0.995 -> 0.993 | Hurt |
| Relax orthogonality constraints and use random initialization instead of spectral initialization | DPCP for unknown-codimension subspace recovery · Washington DC Mall AVIRIS HSI, unknown codimension | F1 score: 0 -> 0.994 (r = 80%); 0 -> 0.993 (r = 90%) | Helped |
Closest papers by meaning
- Dual Principal Component Pursuit for Robust Subspace Learning: Theory and Algorithms for a Holistic Approach (ICML 2021 · similarity 0.94)
- Dual Principal Component Pursuit: Improved Analysis and Efficient Algorithms (NeurIPS 2018 · similarity 0.91)
- Noisy Dual Principal Component Pursuit (ICML 2019 · similarity 0.90)
- Square Root Principal Component Pursuit: Tuning-Free Noisy Robust Matrix Recovery (NeurIPS 2021 · similarity 0.89)
- Low-Rank-Sparse Subspace Representation for Robust Regression (CVPR 2017 · similarity 0.89)
- A Subspace-Constrained Tyler's Estimator and its Applications to Structure from Motion (CVPR 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.