Near-optimal stochastic approximation for online principal component estimation

DOI10.1007/s10107-017-1182-zzbMath1414.62214arXiv1603.05305OpenAlexW2298720961MaRDI QIDQ681490

Tong Zhang, Chris Junchi Li, Mengdi Wang, Han Liu

Publication date: 12 February 2018

Published in: Mathematical Programming. Series A. Series B (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1603.05305

zbMATH Keywords

stochastic approximationprincipal component analysisnonconvex optimizationhigh-dimensional dataonline algorithmstochastic gradient methodfinite-sample analysis

Mathematics Subject Classification ID

Factor analysis and principal components; correspondence analysis (62H25) Stochastic ordinary differential equations (aspects of stochastic analysis) (60H10) Stochastic partial differential equations (aspects of stochastic analysis) (60H15) Online algorithms; streaming algorithms (68W27)

Related Items

Convergence analysis of Oja's iteration for solving online PCA with nonzero-mean samples, Error Estimates for Euler Discretization of High-Index Saddle Dynamics, Nearly optimal stochastic approximation for online principal subspace estimation, Stochastic Gauss-Newton algorithms for online PCA, On the optimality of the Oja's algorithm for online PCA, A brief introduction to manifold optimization

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• The Adaptive Lasso and Its Oracle Properties
• Sparse principal component analysis and iterative thresholding
• Stochastic compositional gradient descent: algorithms for minimizing compositions of expected-value functions
• Random algorithms for convex minimization problems
• Incremental proximal methods for large scale convex optimization
• High-dimensional analysis of semidefinite relaxations for sparse principal components
• Finite sample approximation results for principal component analysis: A matrix perturbation approach
• On stochastic approximation of the eigenvectors and eigenvalues of the expectation of a random matrix
• A simplified neuron model as a principal component analyzer
• Weak convergence and empirical processes. With applications to statistics
• Minimax sparse principal subspace estimation in high dimensions
• Sparse PCA: optimal rates and adaptive estimation
• Incremental Subgradient Methods for Nondifferentiable Optimization
• Stochastic First-Order Methods with Random Constraint Projection
• Estimating the Largest Eigenvalue by the Power and Lanczos Algorithms with a Random Start
• Stability of the Lanczos Method for Matrix Function Approximation
• On Consistency and Sparsity for Principal Components Analysis in High Dimensions
• Information-Theoretic Lower Bounds on the Oracle Complexity of Stochastic Convex Optimization
• Probability