Sparse PCA: optimal rates and adaptive estimation
From MaRDI portal
Publication:2443213
Abstract: Principal component analysis (PCA) is one of the most commonly used statistical procedures with a wide range of applications. This paper considers both minimax and adaptive estimation of the principal subspace in the high dimensional setting. Under mild technical conditions, we first establish the optimal rates of convergence for estimating the principal subspace which are sharp with respect to all the parameters, thus providing a complete characterization of the difficulty of the estimation problem in term of the convergence rate. The lower bound is obtained by calculating the local metric entropy and an application of Fano's lemma. The rate optimal estimator is constructed using aggregation, which, however, might not be computationally feasible. We then introduce an adaptive procedure for estimating the principal subspace which is fully data driven and can be computed efficiently. It is shown that the estimator attains the optimal rates of convergence simultaneously over a large collection of the parameter spaces. A key idea in our construction is a reduction scheme which reduces the sparse PCA problem to a high-dimensional multivariate regression problem. This method is potentially also useful for other related problems.
Recommendations
- Rate-optimal posterior contraction for sparse PCA
- Sparse CCA: adaptive estimation and computational barriers
- Certifiably optimal sparse principal component analysis
- Estimating sparse precision matrix: optimal rates of convergence and adaptive estimation
- Sparse SIR: optimal rates and adaptive estimation
- Sparsistency and agnostic inference in sparse PCA
- Statistical and computational trade-offs in estimation of sparse principal components
- The sparse principal component analysis problem: optimality conditions and algorithms
- Optimal rates of convergence for sparse covariance matrix estimation
- Sparse PCA: convex relaxations, algorithms and applications
Cites work
- A Direct Formulation for Sparse PCA Using Semidefinite Programming
- Adapting to unknown sparsity by controlling the false discovery rate
- Adaptive covariance matrix estimation through block thresholding
- Adaptive estimation of a quadratic functional by model selection.
- Approximation dans les espaces m�triques et th�orie de l'estimation
- Arbitrage, Factor Structure, and Mean-Variance Analysis on Large Asset Markets
- Asymptotics of sample eigenstructure for a large dimensional spiked covariance model
- Combining different procedures for adaptive regression
- Consistency of sparse PCA in high dimension, low sample size contexts
- Convergence of estimates under dimensionality restrictions
- Covariance regularization by thresholding
- Eigenvalues of large sample covariance matrices of spiked population models
- Elements of Information Theory
- Empirical processes with a bounded \(\psi_1\) diameter
- Estimation of (near) low-rank matrices with noise and high-dimensional scaling
- Estimation of high-dimensional low-rank matrices
- Exponential screening and optimal rates of sparse estimation
- Finite sample approximation results for principal component analysis: A matrix perturbation approach
- Functional aggregation for nonparametric regression.
- Gauss-Markov Estimation for Multivariate Linear Models: A Coordinate Free Approach
- Gaussian model selection
- Generalized power method for sparse principal component analysis
- Guaranteed minimum-rank solutions of linear matrix equations via nuclear norm minimization
- High-dimensional analysis of semidefinite relaxations for sparse principal components
- scientific article; zbMATH DE number 3834852 (Why is no real title available?)
- scientific article; zbMATH DE number 3122730 (Why is no real title available?)
- scientific article; zbMATH DE number 3147888 (Why is no real title available?)
- scientific article; zbMATH DE number 47926 (Why is no real title available?)
- scientific article; zbMATH DE number 1964693 (Why is no real title available?)
- scientific article; zbMATH DE number 1522808 (Why is no real title available?)
- Information-theoretic determination of minimax rates of convergence
- Introduction to nonparametric estimation
- Learning Theory
- Minimax bounds for sparse PCA with noisy high-dimensional data
- Nuclear-norm penalization and optimal rates for noisy low-rank matrix completion
- On consistency and sparsity for principal components analysis in high dimensions
- On the distribution of the largest eigenvalue in principal components analysis
- Optimal detection of sparse principal components in high dimension
- Optimal estimation and rank detection for sparse spiked covariance matrices
- Optimal rates of convergence for covariance matrix estimation
- Optimal rates of convergence for sparse covariance matrix estimation
- Oracle inequalities and optimal inference under group sparsity
- PCA consistency in high dimension, low sample size context
- Perturbation bounds in connection with singular value decomposition
- Sparse PCA: optimal rates and adaptive estimation
- Sparse principal component analysis and iterative thresholding
- Sparse principal component analysis via regularized low rank matrix approximation
- Sparse principal component analysis with missing observations
- Sparse Variable PCA Using Geodesic Steepest Descent
- The Rotation of Eigenvectors by a Perturbation. III
- Thresholding for weighted ^2
- Truncated power method for sparse eigenvalue problems
Cited in
(only showing first 100 items - show all)- Sparse Generalized Eigenvalue Problem: Optimal Statistical Rates via Truncated Rayleigh Flow
- Sparse principal component analysis via variable projection
- Bayesian inference for spectral projectors of the covariance matrix
- Testing the order of a population spectral distribution for high-dimensional data
- On two-sample mean tests under spiked covariances
- Sparse principal component regression with adaptive loading
- Sparse CCA: adaptive estimation and computational barriers
- Heterogeneity adjustment with applications to graphical model inference
- The spectral norm of random inner-product kernel matrices
- Rate-optimal perturbation bounds for singular subspaces with applications to high-dimensional statistics
- Detecting Markov random fields hidden in white noise
- Projection tests for high-dimensional spiked covariance matrices
- Robust covariance and scatter matrix estimation under Huber's contamination model
- Optimality and sub-optimality of PCA. I: Spiked random matrix models
- Minimax rates in network analysis: graphon estimation, community detection and hypothesis testing
- Subspace estimation from unbalanced and incomplete data matrices: \({\ell_{2,\infty}}\) statistical guarantees
- Principal component analysis for multivariate extremes
- Bridging convex and nonconvex optimization in robust PCA: noise, outliers and missing data
- Spectral thresholding for the estimation of Markov chain transition operators
- Randomly initialized EM algorithm for two-component Gaussian mixture achieves near optimality in \(O(\sqrt{n})\) iterations
- Testing equivalence of clustering
- A literature review of (Sparse) exponential family PCA
- Estimation of conditional mean operator under the bandable covariance structure
- A high-dimensional test on linear hypothesis of means under a low-dimensional factor model
- Lower bounds for invariant statistical models with applications to principal component analysis
- Sparse SIR: optimal rates and adaptive estimation
- Efficient estimation of linear functionals of principal components
- Consistent estimation of high-dimensional factor models when the factor number is over-estimated
- Convergence of eigenvector empirical spectral distribution of sample covariance matrices
- Nonasymptotic upper bounds for the reconstruction error of PCA
- Limiting laws for divergent spiked eigenvalues and largest nonspiked eigenvalue of sample covariance matrices
- Statistical and computational limits for sparse matrix detection
- From simple structure to sparse components: a review
- Convergence rate of Krasulina estimator
- Subspace perspective on canonical correlation analysis: dimension reduction and minimax rates
- Distributed estimation of principal eigenspaces
- Recovery of simultaneous low rank and two-way sparse coefficient matrices, a nonconvex approach
- Principal component analysis in the local differential privacy model
- Compressed covariance estimation with automated dimension learning
- Provable accelerated gradient method for nonconvex low rank optimization
- Sparse principal component analysis with missing observations
- The two-to-infinity norm and singular subspace geometry with applications to high-dimensional statistics
- ROP: matrix recovery via rank-one projections
- Sparsistency and agnostic inference in sparse PCA
- Optimal estimation and rank detection for sparse spiked covariance matrices
- Rate-optimal posterior contraction for sparse PCA
- Do semidefinite relaxations solve sparse PCA up to the information limit?
- Large covariance estimation through elliptical factor models
- Convergence rates of eigenvector empirical spectral distribution of large dimensional sample covariance matrix
- Minimax sparse principal subspace estimation in high dimensions
- Sparse PCA: optimal rates and adaptive estimation
- Random matrix theory in statistics: a review
- On the optimality of sliced inverse regression in high dimensions
- Optimal estimation for lower bound of the packing number
- Factor models with local factors -- determining the number of relevant factors
- An inexact Riemannian proximal gradient method
- Sparse principal component analysis with missing observations
- Principal component analysis in very high-dimensional spaces
- Robust _1 approaches to computing the geometric median and principal and independent components
- Sparse PCA-based on high-dimensional Itô processes with measurement errors
- Statistical and computational trade-offs in estimation of sparse principal components
- Sparse principal component analysis and iterative thresholding
- Minimax bounds for sparse PCA with noisy high-dimensional data
- Optimal detection of sparse principal components in high dimension
- Near-optimal estimation of simultaneously sparse and low-rank matrices from nested linear measurements
- Recovering PCA and sparse PCA via hybrid-(_1,_2) sparse sampling of data elements
- Scale-invariant sparse PCA on high-dimensional meta-elliptical data
- scientific article; zbMATH DE number 7370563 (Why is no real title available?)
- Exact penalty function for \(\ell_{2,1}\) norm minimization over the Stiefel manifold
- Fast deflation sparse principal component analysis via subspace projections
- Uniform Bounds for Invariant Subspace Perturbations
- Improved shrinkage prediction under a spiked covariance structure
- Posterior contraction rates of the phylogenetic Indian buffet processes
- Optimal sparse singular value decomposition for high-dimensional high-order data
- Perturbation bounds for eigenspaces under a relative gap condition
- Scalable interpretable multi-response regression via SEED
- On consistency and sparsity for principal components analysis in high dimensions
- Wald Statistics in high-dimensional PCA
- Optimally Weighted PCA for High-Dimensional Heteroscedastic Data
- Projected principal component analysis in factor models
- Estimating structured high-dimensional covariance and precision matrices: optimal rates and adaptive estimation
- Solving sparse principal component analysis with global support
- Optimal Permutation Recovery in Permuted Monotone Matrix Model
- Free Energy Wells and Overlap Gap Property in Sparse PCA
- Eigen Selection in Spectral Clustering: A Theory-Guided Practice
- Large volatility matrix analysis using global and national factor models
- Distributed Estimation for Principal Component Analysis: An Enlarged Eigenspace Analysis
- Integrative Factor Regression and Its Inference for Multimodal Data Analysis
- Re-thinking high-dimensional mathematical statistics. Abstracts from the workshop held May 15--21, 2022
- Van Trees inequality, group equivariance, and estimation of principal subspaces
- A penalty-free infeasible approach for a class of nonsmooth optimization problems over the Stiefel manifold
- Bayesian sparse spiked covariance model with a continuous matrix shrinkage prior
- Sparse principal component analysis for high‐dimensional stationary time series
- Euclidean Representation of Low-Rank Matrices and Its Geometric Properties
- Inference for low-rank models
- Long random matrices and tensor unfolding
- Envelopes and principal component regression
- Linearized maximum rank correlation estimation when covariates are functional
- Feature grouping and sparse principal component analysis with truncated regularization
- Exactly Uncorrelated Sparse Principal Component Analysis
This page was built for publication: Sparse PCA: optimal rates and adaptive estimation
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2443213)
