Algorithm 971
From MaRDI portal
Publication:3176327
DOI10.1145/3004053zbMATH Open1391.65085DBLPjournals/toms/LiLSSKT17arXiv1412.3510OpenAlexW2569261326WikidataQ42284279 ScholiaQ42284279MaRDI QIDQ3176327FDOQ3176327
Arthur Szlam, Yuval Kluger, Kelly P. Stanton, George C. Linderman, Huamin Li, Mark Tygert
Publication date: 20 July 2018
Published in: ACM Transactions on Mathematical Software (Search for Journal in Brave)
Abstract: Recent years have witnessed intense development of randomized methods for low-rank approximation. These methods target principal component analysis (PCA) and the calculation of truncated singular value decompositions (SVD). The present paper presents an essentially black-box, fool-proof implementation for Mathworks' MATLAB, a popular software platform for numerical computation. As illustrated via several tests, the randomized algorithms for low-rank approximation outperform or at least match the classical techniques (such as Lanczos iterations) in basically all respects: accuracy, computational efficiency (both speed and memory usage), ease-of-use, parallelizability, and reliability. However, the classical procedures remain the methods of choice for estimating spectral norms, and are far superior for calculating the least singular values and corresponding singular vectors (or singular subspaces).
Full work available at URL: https://arxiv.org/abs/1412.3510
Recommendations
- A randomized algorithm for principal component analysis
- scientific article; zbMATH DE number 839297
- An algorithm for the principal component analysis of large data sets
- Randomized algorithms for distributed computation of principal component analysis and singular value decomposition
- A principal component analysis algorithm with invariant norm
- Principal components: a descent algorithm
- Acceleration of the alternating least squares algorithm for principal components analysis
- Fast algorithms for robust principal component analysis with an upper bound on the rank
- Randomized algorithms for orthogonal nonnegative matrix factorization
Cited In (23)
- Fixed-precision randomized low-rank approximation methods for nonlinear model order reduction of large systems
- Practical Sketching Algorithms for Low-Rank Matrix Approximation
- Efficient Randomized Algorithms for the Fixed-Precision Low-Rank Matrix Approximation
- Optimization of selective phenotyping and population design for genomic prediction
- Randomized Local Model Order Reduction
- A randomized algorithm for tensor singular value decomposition using an arbitrary number of passes
- Single-pass Nyström approximation in mixed precision
- Best Low-rank Approximations and Kolmogorov $n$-widths
- Randomized numerical linear algebra: Foundations and algorithms
- Efficient Error and Variance Estimation for Randomized Matrix Computations
- Randomized LU decomposition using sparse projections
- Randomized near-neighbor graphs, giant components and applications in data science
- Randomized Low-Rank Approximation for Symmetric Indefinite Matrices
- Randomized algorithms for distributed computation of principal component analysis and singular value decomposition
- Fast and Accurate Proper Orthogonal Decomposition using Efficient Sampling and Iterative Techniques for Singular Value Decomposition
- Pass-efficient randomized LU algorithms for computing low-rank matrix approximation
- Scalable Semidefinite Programming
- XT<scp>race</scp>: Making the Most of Every Sample in Stochastic Trace Estimation
- Learning to Forecast Dynamical Systems from Streaming Data
- Pass-Efficient Randomized Algorithms for Low-Rank Matrix Approximation Using Any Number of Views
- Randomized Nyström Preconditioning
- Title not available (Why is that?)
- Randomized model order reduction
Uses Software
This page was built for publication: Algorithm 971
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3176327)
