Research map / Learning theory and algorithms / Matrix and tensor methods
Dual Principal Component Pursuit: Improved Analysis and Efficient Algorithms
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 instead of RANSAC at equal time budget | DPCP-PSGM · KITTI road plane detection | area under ROC: 0.596 -> 0.998 (cloud 1) | Helped |
| Compare DPCP-PSGM with RANSAC given 100x time budget | DPCP-PSGM · KITTI road plane detection | area under ROC: 0.868 vs 0.965 (cloud 120) | Mixed |
| Use DPCP-PSGM instead of REAPER | DPCP-PSGM · KITTI road plane detection | area under ROC: 0.868 vs 0.839 (cloud 120); 0.749 vs 0.749 (cloud 153) | No clear effect |
| Use DPCP-PSGM instead of l2,1-RPCA | DPCP-PSGM · KITTI road plane detection | area under ROC: 0.837 -> 0.991 (cloud 21) | Helped |
| Use piecewise geometrically diminishing step size instead of constant or classical diminishing step size (ablation) | PSGM · synthetic hyperplane, D=30, N=500, 70% outliers | convergence speed | Helped |
| Use PSGM instead of LP-based ALP (ablation) | DPCP · synthetic hyperplane, D=30, N=500 | computing time | Helped |
| Use PSGM-PGD instead of IRLS (ablation) | DPCP · synthetic hyperplane, D=30, N=500 | computing time | Helped |
Closest papers by meaning
- Noisy Dual Principal Component Pursuit (ICML 2019 · similarity 0.95)
- Dual Principal Component Pursuit for Robust Subspace Learning: Theory and Algorithms for a Holistic Approach (ICML 2021 · similarity 0.93)
- Implicit Bias of Projected Subgradient Method Gives Provable Robust Recovery of Subspaces of Unknown Codimension (ICLR 2022 · similarity 0.91)
- Rethinking PCA Through Duality (NeurIPS 2025 · similarity 0.86)
- Fun with Flags: Robust Principal Directions via Flag Manifolds (CVPR 2024 · similarity 0.85)
- Efficient Sparse PCA via Block-Diagonalization (ICLR 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.