Randomized QR with Column Pivoting

Abstract

The dominant contribution to communication complexity in factorizing a matrix using QR with column pivoting is due to column-norm updates that are required to process pivot decisions. We use randomized sampling to approximate this process which dramatically reduces communication in column selection. We also introduce a sample update formula to reduce the cost of sampling trailing matrices. Using our column selection mechanism we observe results that are comparable in quality to those obtained from the QRCP algorithm, but with performance near unpivoted QR. We also demonstrate strong parallel scalability on shared memory multiple core systems using an implementation in Fortran with OpenMP. This work immediately extends to produce low-rank truncated approximations of large matrices. We propose a truncated QR factorization with column pivoting that avoids trailing matrix updates which are used in current implementations of level-3 BLAS QR and QRCP. Provided the truncation rank is small, avoiding trailing matrix updates reduces approximation time by nearly half. By using these techniques and employing a variation on Stewart's QLP algorithm, we develop an approximate truncated SVD that runs nearly as fast as truncated QR.

References in corpus (1)

• Blocked rank-revealing QR factorizations: How randomized sampling can be used to avoid single-vector pivoting

Cited by in corpus (19)

• Data-Driven Sparse Sensor Placement for Reconstruction
• Randomized Matrix Decompositions using R
• Predicting shim gaps in aircraft assembly with machine learning and sparse sensing
• Randomized Local Model Order Reduction
• RSVDPACK: An implementation of randomized algorithms for computing the singular value, interpolative, and CUR decompositions of matrices on multi-core and GPU architectures
• Fast Parallel Randomized QR with Column Pivoting Algorithms for Reliable Low-rank Matrix Approximations
• Randomized Numerical Linear Algebra: Foundations & Algorithms
• Projection-based QLP Algorithm for Efficiently Computing Low-Rank Approximation of Matrices
• Efficient GPU implementation of randomized SVD and its applications
• A Training Set Subsampling Strategy for the Reduced Basis Method
• Deviation Maximization for Rank-Revealing QR Factorizations
• A Generalized Randomized Rank-Revealing Factorization
• PySensors: A Python Package for Sparse Sensor Placement
• Error Estimation for Sketched SVD via the Bootstrap
• Randomized Complete Pivoting for Solving Symmetric Indefinite Linear Systems
• Compressed Randomized UTV Decompositions for Low-Rank Approximations and Big Data Applications
• Randomized Projection for Rank-Revealing Matrix Factorizations and Low-Rank Approximations
• Study of Compressed Randomized UTV Decompositions for Low-Rank Matrix Approximations in Data Science
• Randomized Rank-Revealing UZV Decomposition for Low-Rank Approximation of Matrices