AI Paper Atlas

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

Two-way kernel matrix puncturing: towards resource-efficient PCA and spectral clustering

Romain Couillet, Florent Chatelain, Nicolas Le Bihan

ICML 2021 · Learning theory and algorithms › Matrix and tensor methods · approach: Randomized sketching for matrices · 3 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 data-entry puncturing instead of Gram-entry puncturing at equal sparsity costTwo-way puncturing · Synthetic two-class Gaussian mixture, n=4,000, p=2,000dominant eigenvector structureNo clear effect
Increase n fourfoldTwo-way puncturing on BigGAN-generated VGG features · BigGAN-generated VGG features, tabby cats vs collie dogsclustering performance transition εB: .05 to .015Improved
Puncture both data entries and kernel entriesTwo-way puncturing · BigGAN-generated VGG features and Fashion-MNISTclustering performanceMixed
Keep εS²εB constant while reducing εBTwo-way puncturing · BigGAN-generated VGG features, tabby cats vs collie dogsclustering performanceImproved
Use intensive puncturing with small εB and εSTwo-way puncturing · BigGAN-generated VGG featuresspectrum agreement with predicted uncorrelated-vector limitImproved

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