AI Paper Atlas

Research map / Learning theory and algorithms / Matrix and tensor methods

Bilinear Factor Matrix Norm Minimization for Robust PCA: Algorithms and Applications

Fanhua Shang, James Sheung-Chak Cheng, Yuanyuan Liu, Zhi-Quan Tom Luo, Zhouchen Lin

TPAMI 2017 · Learning theory and algorithms › Matrix and tensor methods · approach: Sparse and robust PCA · 175 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (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 changedWhereMeasuredEffect
use bilinear factor norm penalties (S+L)_2/3 instead of convex nuclear norm RPCARPCA · Corrupted synthetic matrices (noise 0.5, 20% outliers) 500x500RSE: 0.1265 -> 0.0453Improved
use (S+L)_2/3 instead of Unifying (bilinear spectral penalty)Unifying · Corrupted synthetic matrices (noise 0.5, 20% outliers) 1000x1000RSE: 0.0443 -> 0.0318Improved
use (S+L)_2/3 instead of WNNM for speedWNNM · Corrupted synthetic matrices (noise 0.5, 20% outliers) 5000x5000time (s): 24,370.85 -> 128.11Improved
use (S+L)_2/3 instead of RPCA on 10,000x10,000 matricesRPCA · Corrupted synthetic matrices (noise 0.5, 20% outliers) 10000x10000time (s): 20,321.36 -> 487.46Improved
use (S+L)_1/2 vs (S+L)_2/3 on corrupted matrices(S+L)_1/2 · Corrupted synthetic matrices (noise 0.5, 20% outliers) 500x500RSE: 0.0469 -> 0.0453Hurt
add low-rank regularizer (bilinear norm penalties) to plain matrix factorization LMaFit (ablation)LMaFit · Corrupted synthetic matrices (noise 0.5, 20% outliers) 1000x1000, noise 0.2RSE and F-measureImproved
use bilinear norm penalties instead of Schatten quasi-norm LpSq at high missing ratioLpSq · Corrupted matrices, 80% missing, 5% outliersRSEImproved
use bilinear norm penalties instead of LpSq with sufficient observationsLpSq · Corrupted matrices, low missing ratioRSENo clear effect
use (S+L)_1/2 instead of LpSq for text removalLpSq · Text removal (256x256)RSE: 0.1097 -> 0.0396Improved
use (S+L) methods instead of LpSq for text removal speedLpSq · Text removal (256x256)time (s): 77.65 -> 1.21Improved
use D-N/F-N penalties instead of nuclear norm solvers for inpaintingnuclear norm solvers · Image inpainting (85% missing)PSNRImproved
use our methods instead of RegL1 for moving object detectionRegL1 · Surveillance videos (5 sequences)F-measureImproved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy