Nuclear-norm penalization and optimal rates for noisy low-rank matrix completion

DOI10.1214/11-AOS894zbMath1231.62097arXiv1011.6256OpenAlexW2962769133WikidataQ105584275 ScholiaQ105584275MaRDI QIDQ661157

Karim Lounici, Alexandre B. Tsybakov, Vladimir I. Koltchinskii

Publication date: 21 February 2012

Published in: The Annals of Statistics (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1011.6256

zbMATH Keywords

optimal rate of convergence statistical learning lasso low-rank matrix estimation noncommutative Bernstein inequality recovery of the rank

Mathematics Subject Classification ID

Estimation in multivariate analysis (62H12) Gaussian processes (60G15) Inequalities; stochastic orderings (60E15) Linear inference, regression (62J99) Random matrices (probabilistic aspects) (60B20)

Related Items (only showing first 100 items - show all)

Matrix Completion Methods for Causal Panel Data Models ⋮ An Exact and Robust Conformal Inference Method for Counterfactual and Synthetic Controls ⋮ Matrix Completion under Low-Rank Missing Mechanism ⋮ Low Permutation-rank Matrices: Structural Properties and Noisy Completion ⋮ Lifting for Blind Deconvolution in Random Mask Imaging: Identifiability and Convex Relaxation ⋮ Unnamed Item ⋮ Deterministic algorithms for matrix completion ⋮ Unnamed Item ⋮ Unnamed Item ⋮ Statistically optimal and computationally efficient low rank tensor completion from noisy entries ⋮ Bayesian singular value regularization via a cumulative shrinkage process ⋮ Sample average approximation with heavier tails II: localization in stochastic convex optimization and persistence results for the Lasso ⋮ Multistage Convex Relaxation Approach to Rank Regularized Minimization Problems Based on Equivalent Mathematical Program with a Generalized Complementarity Constraint ⋮ Matrix completion under complex survey sampling ⋮ High-dimensional latent panel quantile regression with an application to asset pricing ⋮ Oracle inequality for sparse trace regression models with exponential \(\beta\)-mixing errors ⋮ The rate of convergence for sparse and low-rank quantile trace regression ⋮ Robust Recommendation via Social Network Enhanced Matrix Completion ⋮ Sparse and Low-Rank Matrix Quantile Estimation With Application to Quadratic Regression ⋮ A framework of regularized low-rank matrix models for regression and classification ⋮ Learning Markov Models Via Low-Rank Optimization ⋮ A randomised iterative method for solving factorised linear systems ⋮ Profile GMM estimation of panel data models with interactive fixed effects ⋮ Gradient-Based Markov Chain Monte Carlo for Bayesian Inference With Non-differentiable Priors ⋮ The Lasso with structured design and entropy of (absolute) convex hulls ⋮ Bayesian uncertainty quantification for low-rank matrix completion ⋮ Covariate-assisted matrix completion with multiple structural breaks ⋮ A portmanteau local feature discrimination approach to the classification with high-dimensional matrix-variate data ⋮ Robust matrix estimations meet Frank-Wolfe algorithm ⋮ Low-Rank Regression Models for Multiple Binary Responses and their Applications to Cancer Cell-Line Encyclopedia Data ⋮ High-dimensional estimation of quadratic variation based on penalized realized variance ⋮ High-dimensional VARs with common factors ⋮ Inference for low-rank models ⋮ Entrywise limit theorems for eigenvectors of signal-plus-noise matrix models with weak signals ⋮ Expectile trace regression via low-rank and group sparsity regularization ⋮ Inference for low-rank completion without sample splitting with application to treatment effect estimation ⋮ High-dimensional estimation with geometric constraints: Table 1. ⋮ Near-optimal estimation of simultaneously sparse and low-rank matrices from nested linear measurements ⋮ Factor Models for High-Dimensional Tensor Time Series ⋮ On a unified view of nullspace-type conditions for recoveries associated with general sparsity structures ⋮ Asymptotic equivalence of quantum state tomography and noisy matrix completion ⋮ Sparse PCA: optimal rates and adaptive estimation ⋮ Noisy low-rank matrix completion with general sampling distribution ⋮ On the finite-sample analysis of \(\Theta\)-estimators ⋮ Noisy Matrix Completion: Understanding Statistical Guarantees for Convex Relaxation via Nonconvex Optimization ⋮ Oracle Inequalities for Local and Global Empirical Risk Minimizers ⋮ On the finite-sample analysis of \(\Theta\)-estimators ⋮ Unnamed Item ⋮ Matrix completion based on Gaussian parameterized belief propagation ⋮ Matrix Denoising for Weighted Loss Functions and Heterogeneous Signals ⋮ On tight bounds for the Lasso ⋮ Confidence Intervals for Low Dimensional Parameters in High Dimensional Linear Models ⋮ Sharp variable selection of a sparse submatrix in a high-dimensional noisy matrix ⋮ Projected principal component analysis in factor models ⋮ High-dimensional regression with unknown variance ⋮ Sparse estimation by exponential weighting ⋮ Discussion of ``Estimating structured high-dimensional covariance and precision matrices: optimal rates and adaptive estimation ⋮ FACTORISABLE MULTITASK QUANTILE REGRESSION ⋮ Intelligent Initialization and Adaptive Thresholding for Iterative Matrix Completion: Some Statistical and Algorithmic Theory forAdaptive-Impute ⋮ Matrix Completion With Covariate Information ⋮ Spectral Methods for Passive Imaging: Nonasymptotic Performance and Robustness ⋮ Sharp Oracle Inequalities for Square Root Regularization ⋮ Regularization and the small-ball method II: complexity dependent error rates ⋮ Low Rank Estimation of Similarities on Graphs ⋮ On the properties of variational approximations of Gibbs posteriors ⋮ Unnamed Item ⋮ Operator Splitting for a Homogeneous Embedding of the Linear Complementarity Problem ⋮ Estimation of a sparse and spiked covariance matrix ⋮ Iterative Methods for Solving Factorized Linear Systems ⋮ Unnamed Item ⋮ ISLET: Fast and Optimal Low-Rank Tensor Regression via Importance Sketching ⋮ An optimal statistical and computational framework for generalized tensor estimation ⋮ Adaptive estimation of the copula correlation matrix for semiparametric elliptical copulas ⋮ Worst possible sub-directions in high-dimensional models ⋮ A shrinkage principle for heavy-tailed data: high-dimensional robust low-rank matrix recovery ⋮ On the prediction loss of the Lasso in the partially labeled setting ⋮ Convex optimization learning of faithful Euclidean distance representations in nonlinear dimensionality reduction ⋮ Ridge-type regularization method for questionnaire data analysis ⋮ Low-rank diffusion matrix estimation for high-dimensional time-changed Lévy processes ⋮ Optimal large-scale quantum state tomography with Pauli measurements ⋮ Generalized co-sparse factor regression ⋮ Inference for low-rank tensors -- no need to debias ⋮ Aggregated hold out for sparse linear regression with a robust loss function ⋮ A distributed Frank-Wolfe framework for learning low-rank matrices with the trace norm ⋮ Tight risk bound for high dimensional time series completion ⋮ Matrix completion via max-norm constrained optimization ⋮ Geometric inference for general high-dimensional linear inverse problems ⋮ A rank-corrected procedure for matrix completion with fixed basis coefficients ⋮ Estimation of low rank density matrices: bounds in Schatten norms and other distances ⋮ Estimation of matrices with row sparsity ⋮ Sharp MSE bounds for proximal denoising ⋮ A graphical approach to the analysis of matrix completion ⋮ Estimation of low-rank covariance function ⋮ Trace regression model with simultaneously low rank and row(column) sparse parameter ⋮ Low rank estimation of smooth kernels on graphs ⋮ How can we identify the sparsity structure pattern of high-dimensional data: an elementary statistical analysis to interpretable machine learning ⋮ Optimal prediction in the linearly transformed spiked model ⋮ Double instrumental variable estimation of interaction models with big data ⋮ Matrix completion by singular value thresholding: sharp bounds ⋮ Bayesian linear regression with sparse priors

Cites Work

This page was built for publication: Nuclear-norm penalization and optimal rates for noisy low-rank matrix completion