AI Paper Atlas

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

Robust High-Dimensional Low-Rank Matrix Estimation: Optimal Rate and Data-Adaptive Tuning

Xiaolong Cui, Lei Shi, Wei Zhong, Changliang Zou

JMLR 2023 · 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 rank-based Wilcoxon loss instead of least-squares loss in nuclear-norm estimationmatrix lasso · matrix regression, Cauchy/log-normal errorslog Frobenius errorImproved
use rank-based loss versus l1 (LAD) loss under Gaussian errorsRML · matrix regression, Gaussian errorslog Frobenius errorImproved
use RML instead of robustified matrix lasso with robust CV tuning, Cauchy errorsmatrix regression · Table 1 (I) Cauchysquared Frobenius error: 0.105 -> <0.001Improved
use RML instead of robustified matrix lasso, Normal errorsmatrix regression · Table 1 (I) Normalsquared Frobenius error: 0.114 -> 0.130Hurt
use RML instead of robustified matrix lasso, multivariate regression Cauchymultivariate regression · Table 1 (II) Cauchysquared Frobenius error: 0.076 -> 0.003Improved
use RML instead of robustified matrix lasso, matrix completion log-normalmatrix completion · Table 1 (III) Log-normalsquared Frobenius error: 0.121 -> 0.048Improved
pivotal tuning of RML instead of robust cross-validation, tuning timematrix regression · Table 1 (I) Normaltotal time (s): 344 -> 1.97Improved
pivotal tuning of RML instead of robust CV, matrix completion total timematrix completion · Table 1 (III) Normaltotal time (s): 370 -> 3.66Improved
use RML instead of robustified matrix lasso for rank recovery under heavy tailsmultivariate regression · Table 1 (II) Log-normalestimated rank: 21.83 -> 5Improved
use RML instead of fixed-rank factorization methods (SGD, SSM) with misspecified rank R=3,10matrix regression · Figure 2log Frobenius errorImproved
increase t-distribution degrees of freedom d with l1 loss versus RML (ablation)multivariate regression · Figure 4Frobenius errorMixed
use RML instead of l2 matrix lasso on gene dataArabidopsis thaliana multivariate regression · Arabidopsis thaliana gene dataMSE: 0.609 -> 0.567Improved

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