Subspace estimation from unbalanced and incomplete data matrices: \({\ell_{2,\infty}}\) statistical guarantees
From MaRDI portal
Publication:2039795
DOI10.1214/20-AOS1986zbMath1471.62371arXiv1910.04267MaRDI QIDQ2039795
H. Vincent Poor, Yuxin Chen, Gen Li, Changxiao Cai, Yuejie Chi
Publication date: 5 July 2021
Published in: The Annals of Statistics (Search for Journal in Brave)
Full work available at URL: https://arxiv.org/abs/1910.04267
principal component analysis; spectral method; missing data; spectral clustering; covariance estimation; tensor completion; leave-one-out analysis
62H25: Factor analysis and principal components; correspondence analysis
62H12: Estimation in multivariate analysis
62H30: Classification and discrimination; cluster analysis (statistical aspects)
62D10: Missing data
Related Items
Iterative Collaborative Filtering for Sparse Matrix Estimation, Nonconvex Low-Rank Tensor Completion from Noisy Data, Biwhitening Reveals the Rank of a Count Matrix, Convex and Nonconvex Optimization Are Both Minimax-Optimal for Noisy Blind Deconvolution Under Random Designs, Community detection on mixture multilayer networks via regularized tensor decomposition, An \({\ell_p}\) theory of PCA and spectral clustering
Cites Work
- Consistency thresholds for the planted bisection model
- On tensor completion via nuclear norm minimization
- Sparse principal component analysis and iterative thresholding
- High-dimensional covariance matrix estimation with missing observations
- Reconstruction and estimation in the planted partition model
- Asymptotics and concentration bounds for bilinear forms of spectral projectors of sample covariance
- Angular synchronization by eigenvectors and semidefinite programming
- Spectral clustering and the high-dimensional stochastic blockmodel
- High-dimensional regression with noisy and missing data: provable guarantees with nonconvexity
- Community detection in sparse networks via Grothendieck's inequality
- Minimax rate-optimal estimation of high-dimensional covariance matrices with incomplete data
- On sample eigenvalues in a generalized spiked population model
- Delocalization of eigenvectors of random matrices with independent entries
- Covariance regularization by thresholding
- Finite sample approximation results for principal component analysis: A matrix perturbation approach
- Weighted least squares fitting using ordinary least squares algorithms
- Random perturbation of low rank matrices: improving classical bounds
- On semidefinite relaxations for the block model
- Random Laplacian matrices and convex relaxations
- On the distribution of the largest eigenvalue in principal components analysis
- Adaptive covariance matrix estimation through block thresholding
- On polynomial time methods for exact low-rank tensor completion
- Implicit regularization in nonconvex statistical estimation: gradient descent converges linearly for phase retrieval, matrix completion, and blind deconvolution
- Entrywise eigenvector analysis of random matrices with low expected rank
- Spectral method and regularized MLE are both optimal for top-\(K\) ranking
- The two-to-infinity norm and singular subspace geometry with applications to high-dimensional statistics
- Matrix estimation by universal singular value thresholding
- Consistency of spectral clustering in stochastic block models
- Optimal estimation and rank detection for sparse spiked covariance matrices
- Robust and computationally feasible community detection in the presence of arbitrary outlier nodes
- Convexified modularity maximization for degree-corrected stochastic block models
- Sparse PCA: optimal rates and adaptive estimation
- Exact matrix completion via convex optimization
- Statistically optimal and computationally efficient low rank tensor completion from noisy entries
- Asymmetry helps: eigenvalue and eigenvector analyses of asymmetrically perturbed low-rank matrices
- Optimal Estimation and Completion of Matrices with Biclustering Structures
- Sparse Principal Component Analysis with Missing Observations
- Perturbation of Linear Forms of Singular Vectors Under Gaussian Noise
- Phase transitions in semidefinite relaxations
- Guaranteed Matrix Completion via Non-Convex Factorization
- Information Recovery From Pairwise Measurements
- Achieving Exact Cluster Recovery Threshold via Semidefinite Programming: Extensions
- Exact Recovery in the Stochastic Block Model
- Tensor completion and low-n-rank tensor recovery via convex optimization
- Oracle Inequalities for Local and Global Empirical Risk Minimizers
- A spectral heuristic for bisecting random graphs
- Graph Partitioning via Adaptive Spectral Techniques
- Community Detection and Stochastic Block Models
- An $\ell_{\infty}$ Eigenvector Perturbation Bound and Its Application to Robust Covariance Estimation
- Tensor SVD: Statistical and Computational Limits
- Incoherent Tensor Norms and Their Applications in Higher Order Tensor Completion
- Sequential Unfolding SVD for Tensors With Applications in Array Signal Processing
- The Projected Power Method: An Efficient Algorithm for Joint Alignment from Pairwise Differences
- Tensor-Based Formulation and Nuclear Norm Regularization for Multienergy Computed Tomography
- A Simple SVD Algorithm for Finding Hidden Partitions
- Asymptotic theory for estimating the singular vectors and values of a partially-observed low rank matrix with noise
- Near-Optimal Bounds for Phase Synchronization
- A Consistent Adjacency Spectral Embedding for Stochastic Blockmodel Graphs
- Spectral Algorithms for Tensor Completion
- Covariance Matrix Estimation With Non Uniform and Data Dependent Missing Observations
- Optimal Bipartite Network Clustering
- Sparse and Low-Rank Tensor Estimation via Cubic Sketchings
- Noisy Matrix Completion: Understanding Statistical Guarantees for Convex Relaxation via Nonconvex Optimization
- Nonconvex Optimization Meets Low-Rank Matrix Factorization: An Overview
- On Consistency and Sparsity for Principal Components Analysis in High Dimensions
- Community detection thresholds and the weak Ramanujan property
- Recovering Low-Rank Matrices From Few Coefficients in Any Basis
- Matrix Completion From a Few Entries
- The Power of Convex Relaxation: Near-Optimal Matrix Completion
- Achieving Optimal Misclassification Proportion in Stochastic Block Model
- Fast spectral algorithms from sum-of-squares proofs: tensor decomposition and planted sparse vectors
- The Sup-norm Perturbation of HOSVD and Low Rank Tensor Denoising
- Low-rank matrix completion using alternating minimization
- Perturbation theory for pseudo-inverses
- Signal-plus-noise matrix models: eigenvector deviations and fluctuations
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item