An Alternating Manifold Proximal Gradient Method for Sparse PCA and Sparse CCA
arXiv:1903.11576
Abstract
Sparse principal component analysis (PCA) and sparse canonical correlation analysis (CCA) are two essential techniques from high-dimensional statistics and machine learning for analyzing large-scale data. Both problems can be formulated as an optimization problem with nonsmooth objective and nonconvex constraints. Since non-smoothness and nonconvexity bring numerical difficulties, most algorithms suggested in the literature either solve some relaxations or are heuristic and lack convergence guarantees. In this paper, we propose a new alternating manifold proximal gradient method to solve these two high-dimensional problems and provide a unified convergence analysis. Numerical experiment results are reported to demonstrate the advantages of our algorithm.
References in corpus (9)
- Generalized power method for sparse principal component analysis
- Global rates of convergence for nonconvex optimization on manifolds
- Optimal Solutions for Sparse Principal Component Analysis
- Truncated Power Method for Sparse Eigenvalue Problems
- Complete Dictionary Recovery over the Sphere I: Overview and the Geometric Picture
- Sparse Principal Component Analysis via Variable Projection
- Nonconvex phase synchronization
- A greedy approach to sparse canonical correlation analysis
- Sparse CCA via Precision Adjusted Iterative Thresholding
Cited by in corpus (5)
- Robust Low-rank Matrix Completion via an Alternating Manifold Proximal Gradient Continuation Method
- Riemannian Stochastic Proximal Gradient Methods for Nonsmooth Optimization over the Stiefel Manifold
- Sparse principal component regression via singular value decomposition approach
- A Manifold Proximal Linear Method for Sparse Spectral Clustering with Application to Single-Cell RNA Sequencing Data Analysis
- Multi-Rank Sparse and Functional PCA: Manifold Optimization and Iterative Deflation Techniques