AI Paper Atlas

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

Provably Correct Algorithms for Matrix Column Subset Selection with Selectively Sampled Data

Yining Wang, Aarti Singh

JMLR 2018 · Learning theory and algorithms › Matrix and tensor methods · approach: Low-rank approximation theory

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 iterative norm sampling instead of norm samplingcolumn subset selection with missing entries · Lena image (50 columns, alpha=0.3)relative selection error: 0.106 -> 0.088Improved
use approximate leverage score sampling instead of norm samplingcolumn subset selection with missing entries · Lena image (50 columns, alpha=0.3)relative selection error: 0.106 -> 0.103Improved
use iterative norm sampling or leverage score sampling instead of block OMP on coherent-column matricesBlock OMP · synthetic matrices with coherent columnsselection errorImproved
sample without replacement instead of with replacement (ablation)norm and leverage score sampling · synthetic Gaussian and coherent matricesselection errorImproved
use iterative norm sampling when both target rank and missing rate are highiterative norm sampling · synthetic Gaussian low-rank plus noise matricesselection errorHurt
use iterative norm sampling instead of leverage score sampling on very low-noise near-low-rank matricesleverage score sampling · synthetic low-rank plus noise matricesselection errorImproved
increase number of repeated columns (more coherence) (ablation)norm sampling, block OMP, group Lasso · synthetic noisy low-rank, sigma=0.1, s=k=15selection errorHurt
increase number of repeated columns (more coherence) (ablation)volume sampling and iterative norm sampling · synthetic noisy low-rank, sigma=0.1, s=k=15selection errorNo clear effect

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