Research map / Learning theory and algorithms / Matrix and tensor methods
Learned Robust PCA: A Scalable Deep Unfolding Approach for High-Dimensional Outlier Detection
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 LRPCA (learned parameters) instead of ScaledGD and AltProj | LRPCA · VIRAT video background subtraction | runtime (secs): 260.45 -> 16.01 vs ScaledGD on ParkingLot1 | Helped |
| use LRPCA instead of CORONA | LRPCA · Ultrasound imaging | average inference time (secs): 0.9225 -> 0.0057 | Helped |
| use LRPCA instead of CORONA | LRPCA · Ultrasound imaging | average MSE loss: 4.88 -> 9.97 (x10^-4) | Hurt |
| use LRPCA instead of ScaledGD at high outlier density | LRPCA · Synthetic RPCA, alpha 0.5-0.6 | recovery success out of 10: 0/10 -> 10/10 at alpha=0.5 | Helped |
| use FRMNN (feedforward plus recurrent layers) instead of pure RNN unfolding (ablation) | LRPCA · Synthetic RPCA convergence | convergence error | Helped |
Closest papers by meaning
- RES-PCA: A Scalable Approach to Recovering Low-Rank Matrices (CVPR 2019 · similarity 0.88)
- Robust Formulation for PCA: Avoiding Mean Calculation With L2,p-norm Maximization (AAAI 2018 · similarity 0.88)
- Residual-Based Sampling for Online Outlier-Robust PCA (ICML 2022 · similarity 0.88)
- Robust Principal Component Analysis with Side Information (ICML 2016 · similarity 0.88)
- Nearly-Linear Time and Streaming Algorithms for Outlier-Robust PCA (ICML 2023 · similarity 0.87)
- Capturing the denoising effect of PCA via compression ratio (NeurIPS 2024 · 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.