AI Paper Atlas

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

Power Iteration for Tensor PCA

Jiaoyang Huang, Daniel Z. Huang, Qing Yang, Guang Cheng

JMLR 2022 · Learning theory and algorithms › Matrix and tensor methods · approach: Randomized tensor decomposition algorithms · 12 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
increase signal strength from beta=5 to beta=10rank-one tensor power iteration · Rank-one spiked tensor (n=600, k=3)Gaussian fit of normalized estimators: beta=5 -> beta=10Improved
use informative initialization (v+w)/norm instead of randomtensor power iteration · Rank-one spiked tensor (n=600, k=3)asymptotic Gaussianity at small beta: beta=5, beta=10Improved
increase signal strength beta above n^((k-2)/2) with random initializationtensor power iteration · Rank-one spiked tensor (k=3)signal recoveryImproved
increase informative initialization strength beta<u0,v> with fixed beta=3tensor power iteration · Rank-one spiked tensor (k=3)signal recovery: beta<u0,v> in (0,2]Improved
increase dimension n from 50 to 800tensor power iteration · Rank-2 spiked tensor, k=395% CI coverage for beta1: 0.9136 -> 0.9551Mixed
increase dimension n from 50 to 800tensor power iteration · Rank-2 spiked tensor, k=3fraction p1 converging to beta1: 0.405 -> 0.401No 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