AI Paper Atlas

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

Rate Optimal Denoising of Simultaneously Sparse and Low Rank Matrices

Dan Yang, Zongming Ma, Andreas Buja

JMLR 2016 · Learning theory and algorithms › Matrix and tensor methods · approach: Low-rank approximation theory · 26 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 sparsity block size from k=l=50 to k=100, l=200Iterative thresholding denoiser · Synthetic denoising, m=2000 n=1000 r=10average L2 loss: 1133.03 -> 3598.69Hurt
increase sparsity block size from k=l=50 to k=100, l=200Iterative thresholding denoiser · Synthetic denoising, m=2000 n=1000 r=10average L1 loss: 19056.47 -> 65099.19Hurt
vary sparsity level (k,l) with loss rescaled by theoretical rateIterative thresholding denoiser · Synthetic denoising, m=2000 n=1000 r=10rescaled L2 loss: 0.64, 0.60, 0.68, 0.63No clear effect
vary sparsity level (k,l) with loss rescaled by theoretical rateIterative thresholding denoiser · Synthetic denoising, m=2000 n=1000 r=10rescaled L1 loss: 1.08, 0.98, 1.23, 1.07No clear effect
increase noise standard deviation sigma from 0.2 to 2Iterative thresholding denoiser · Synthetic denoising, r=10, k=l=50L1 and L2 lossHurt
increase rank r from 1 to 10Iterative thresholding denoiser · Synthetic denoising, sigma=1, k=l=50L1 lossHurt
increase rank r from 1 to 10Iterative thresholding denoiser · Synthetic denoising, sigma=1, k=l=50L2 lossHurt
vary loss parameter q between 1 and 2Iterative thresholding denoiser · Synthetic denoising, sigma=1log loss vs 1/qNo clear effect
use data-based rank selection instead of known rankIterative thresholding denoiser · Synthetic denoising, r=10selected rank: r = 10Improved
increase spikiness parameter s from 0.1 to 1.1Iterative thresholding denoiser · Synthetic denoising, r=1, k=l=50, sigma=1L1 and L2 lossHurt
increase spikiness parameter s beyond 1.1 to 1.6Iterative thresholding denoiser · Synthetic denoising, r=1, k=l=50, sigma=1L1 and L2 lossImproved

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