AI Paper Atlas

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

Recovering PCA and Sparse PCA via Hybrid-(l1,l2) Sparse Sampling of Data Elements

Abhisek Kundu, Petros Drineas, Malik Magdon-Ismail

JMLR 2017 · Learning theory and algorithms › Matrix and tensor methods · approach: Sparse and robust PCA · 8 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 hybrid-(l1,l2) sampling instead of l1 samplingsparse sketch for PCA · Stock PCA (2% non-zeros)variance preserved: 85.09 % -> 97.07%Improved
use hybrid-(l1,l2) sampling instead of Bernstein samplingsparse sketch for PCA · Stock PCA (2% non-zeros)variance preserved: 95.61 % -> 97.07%Improved
use hybrid-(l1,l2) sampling instead of l1 or Bernstein samplingsparse sketch for PCA · Digit PCA (9% non-zeros)variance preserved: 98.47 % and 98.46 % vs 98.51%No clear effect
use hybrid sampling PCA instead of Gaussian random projection PCADigit PCA · Digit (7% non-zeros)variance preserved: 96.89% -> 97.94%Improved
use hybrid-(l1,l2) sketch instead of uniform sketch for sparse PCASpasm sparse PCA r=40 · Genef(sp)/f(G sp): 0.15 -> 0.88Improved
use hybrid-(l1,l2) sketch instead of uniform sketch for sparse PCASpasm sparse PCA r=40 · TechTC TTC1f(sp)/f(G sp): 0.38 -> 0.99Improved
use hybrid-(l1,l2) sketch instead of uniform sketch for sparse PCASpasm sparse PCA r=40 · Digitf(max)/f(G max): 1.01 vs 0.99No clear effect
use hybrid-(l1,l2) sketch instead of uniform sketch for sparse PCASpasm sparse PCA r=40 · Stockf(max)/f(G max): 0.66 -> 1.00Improved
use optimal mixing alpha* instead of alpha = 1 (pure l1) (ablation)hybrid-(l1,l2) sketch · Stock sparse PCAf(Hsp,r)Improved
use hybrid-(l1,l2) sampling instead of leverage score samplinglow-rank power-law matrix sparsification · Power-law matrix gamma=1.0, s/(k(m+n))=3spectral-norm error: 42% -> 8%Improved
use hybrid-(l1,l2) sampling instead of leverage score samplingrank-truncated A0.1 sparsification · A0.1, k=5, s/(k(m+n))=3spectral-norm error: 80% -> 25%Improved
use hybrid-(l1,l2) sampling instead of l2 sampling with truncation epsilon=0.1sparse sketch · A0.1 syntheticspectral-norm errorHurt

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