Generalized power method for sparse principal component analysis
From MaRDI portal
Abstract: In this paper we develop a new approach to sparse principal component analysis (sparse PCA). We propose two single-unit and two block optimization formulations of the sparse PCA problem, aimed at extracting a single sparse dominant principal component of a data matrix, or more components at once, respectively. While the initial formulations involve nonconvex functions, and are therefore computationally intractable, we rewrite them into the form of an optimization program involving maximization of a convex function on a compact set. The dimension of the search space is decreased enormously if the data matrix has many more columns (variables) than rows. We then propose and analyze a simple gradient method suited for the task. It appears that our algorithm has best convergence properties in the case when either the objective function or the feasible set are strongly convex, which is the case with our single-unit formulations and can be enforced in the block case. Finally, we demonstrate numerically on a set of random and gene expression test problems that our approach outperforms existing algorithms both in quality of the obtained solution and in computational speed.
Recommendations
- Sparse principal component analysis by choice of norm
- An exact approach to sparse principal component analysis
- Projection sparse principal component analysis: an efficient least squares method
- Sparse PCA: convex relaxations, algorithms and applications
- An augmented Lagrangian approach for sparse principal component analysis
Cited in
(only showing first 100 items - show all)- A majorization-minimization approach to the sparse generalized eigenvalue problem
- Robust sparse principal component analysis
- Global convergence of Riemannian line search methods with a Zhang-Hager-type condition
- Nonsmooth optimization over the Stiefel manifold and beyond: proximal gradient method and recent variants
- Estimation of low-rank matrices via approximate message passing
- Bayesian variable selection for globally sparse probabilistic PCA
- Sparse principal component analysis via variable projection
- Principal Component Analysis by Optimization of Symmetric Functions has no Spurious Local Optima
- Sparse PCA on fixed-rank matrices
- A non-monotone linear search algorithm with mixed direction on Stiefel manifold
- Solving sparse principal component analysis with global support
- Sparsistency and agnostic inference in sparse PCA
- Sparse multivariate functional principal component analysis
- Stochastic proximal gradient method FOR _1 regularized optimization over a sphere
- Solving the k-sparse eigenvalue problem with reinforcement learning
- Comment: Ridge Regression and Regularization of Large Matrices
- Minimax sparse principal subspace estimation in high dimensions
- On the estimation performance and convergence rate of the generalized power method for phase synchronization
- From simple structure to sparse components: a review
- Exactly Uncorrelated Sparse Principal Component Analysis
- Regularized generalized canonical correlation analysis: a framework for sequential multiblock component methods
- Integrative and regularized principal component analysis of multiple sources of data
- Sparse factor analysis for categorical data with the group-sparse generalized singular value decomposition
- On the rotational invariant \(L_1\)-norm PCA
- Certifiably optimal sparse principal component analysis
- Optimal solutions for sparse principal component analysis
- Projected nonmonotone search methods for optimization with orthogonality constraints
- Sparse PCA: convex relaxations, algorithms and applications
- Truncated power method for sparse eigenvalue problems
- A New Basis for Sparse Principal Component Analysis
- A generalized complex power iteration-based algorithm for quartic-quadratic optimization problems over a spherical constraint and its applications
- Semi‐supervised Eigenbasis novelty detection
- Understanding large text corpora via sparse machine learning
- Semi-sparse PCA
- scientific article; zbMATH DE number 7370563 (Why is no real title available?)
- Alternating maximization: unifying framework for 8 sparse PCA formulations and efficient parallel codes
- Sparse power factorization: balancing peakiness and sample complexity
- An efficient algorithm for Fantope-constrained sparse principal subspace estimation problem
- Fast deflation sparse principal component analysis via subspace projections
- Using \(\ell_1\)-relaxation and integer programming to obtain dual bounds for sparse PCA
- Principal minimax support vector machine for sufficient dimension reduction with contaminated data
- Sparse PCA by iterative elimination algorithm
- scientific article; zbMATH DE number 7307468 (Why is no real title available?)
- A Block Lanczos Method for Large-Scale Quadratic Minimization Problems with Orthogonality Constraints
- An \(\ell_1\)-penalized adaptive normalized quasi-Newton algorithm for sparsity-aware generalized eigen-subspace tracking
- Kurdyka-Łojasiewicz property of zero-norm composite functions
- Best subset solution path for linear dimension reduction models using continuous optimization
- Sparse principal component analysis by choice of norm
- A general null space property for sparse principal component analysis
- scientific article; zbMATH DE number 7338724 (Why is no real title available?)
- Riemannian preconditioning
- On conjugate families and Jeffreys priors for von Mises-Fisher distributions
- Orthogonal connectivity factorization: interpretable decomposition of variability in correlation matrices
- Three \(l_1\) based nonconvex methods in constructing sparse mean reverting portfolios
- An active-set proximal quasi-Newton algorithm for ℓ1-regularized minimization over a sphere constraint
- Ellipsoidal embeddings of graphs
- Provable sample-efficient sparse phase retrieval initialized by truncated power method
- scientific article; zbMATH DE number 7625166 (Why is no real title available?)
- Sparse principal component analysis with missing observations
- Sparse PCA: optimal rates and adaptive estimation
- The sparse principal component analysis problem: optimality conditions and algorithms
- Nonmonotone inexact restoration approach for minimization with orthogonality constraints
- Sparse Principal Component Analysis Based on Least Trimmed Squares
- Sparse exponential family principal component analysis
- Polar decomposition-based algorithms on the product of Stiefel manifolds with applications in tensor approximation
- Near-optimal bounds for phase synchronization
- Smart Alpha: active management with unstable and latent factors
- Convexifying the set of matrices of bounded rank: applications to the quasiconvexification and convexification of the rank function
- Cutting plane generation through sparse principal component analysis
- A framework of constraint preserving update schemes for optimization on Stiefel manifold
- Towards statistically provable geometric 3D human pose recovery
- SIMPCA: a framework for rotating and sparsifying principal components
- Inexact primal-dual gradient projection methods for nonlinear optimization on convex set
- The restricted isometry property for random block diagonal matrices
- Nonconvex phase synchronization
- An active-set algorithm for norm constrained quadratic problems
- Fundamental limits of low-rank matrix estimation with diverging aspect ratios
- Sparse principal component analysis subject to prespecified cardinality of loadings
- Sparsifying the least-squares approach to PCA: comparison of lasso and cardinality constraint
- Proximal distance algorithms: theory and practice
- Solving \(\ell_0\)-penalized problems with simple constraints via the Frank-Wolfe reduced dimension method
- PCA Sparsified
- Generalized left-localized Cayley parametrization for optimization with orthogonality constraints
- Sparse and integrative principal component analysis for multiview data
- Robust and sparse pca for high-dimensional data via Huber loss and non-convex regularization
- An exact approach to sparse principal component analysis
- Minimization over the _p ball using a hybrid first-order method
- Maximization of Matrix Trace Function of Product Stiefel Manifolds
- Sparse total least squares: analysis and greedy algorithms
- Implicit steepest descent algorithm for optimization with orthogonality constraints
- Monotonically convergent algorithms for symmetric tensor approximation
- Least squares sparse principal component analysis: a backward elimination approach to attain large loadings
- Identifiability of complete dictionary learning
- Alternating direction method of multipliers for penalized zero-variance discriminant analysis
- Sparse space-time resolvent analysis for statistically stationary and time-varying flows
- Proximal gradient method for nonsmooth optimization over the Stiefel manifold
- ECA: High-Dimensional Elliptical Component Analysis in Non-Gaussian Distributions
- Projection algorithms for nonconvex minimization with application to sparse principal component analysis
- A decentralized proximal gradient tracking algorithm for composite optimization on Riemannian manifolds
- Projection sparse principal component analysis: an efficient least squares method
This page was built for publication: Generalized power method for sparse principal component analysis
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2896039)
