Randomized low-rank approximation of monotone matrix functions

@article{Persson2022RandomizedLA,
  title={Randomized low-rank approximation of monotone matrix functions},
  author={David Persson and Daniel Kressner},
  journal={ArXiv},
  year={2022},
  volume={abs/2209.11023}
}
This work is concerned with computing low-rank approximations of a matrix function f ( A ) for a large symmetric positive semi-definite matrix A , a task that arises in, e.g., statistical learning and inverse problems. The application of popular randomized methods, such as the randomized singular value decomposition or the Nystr¨om approximation, to f ( A ) requires multiplying f ( A ) with a few random vectors. A significant dis-advantage of such an approach, matrix-vector products with f ( A… 

Figures from this paper

References

SHOWING 1-10 OF 49 REFERENCES

Fast and stable randomized low-rank matrix approximation

This work studies a generalization of Nystr{o}m method applicable to general matrices, and shows that it has near-optimal approximation quality comparable to competing methods and can significantly outperform state-of-the-art methods.

Finding Structure with Randomness: Probabilistic Algorithms for Constructing Approximate Matrix Decompositions

This work surveys and extends recent research which demonstrates that randomization offers a powerful tool for performing low-rank matrix approximation, and presents a modular framework for constructing randomized algorithms that compute partial matrix decompositions.

On Randomized Trace Estimates for Indefinite Matrices with an Application to Determinants

New tail bounds for randomized trace estimates applied to indefinite B with Rademacher or Gaussian random vectors are derived, which significantly improve existing results for indefinite B, reducing the number of required samples by a factor n or even more.

Hutch++: Optimal Stochastic Trace Estimation

A new randomized algorithm, Hutch++, is introduced, which computes a (1 ± ε) approximation to tr( A ) for any positive semidefinite (PSD) A using just O(1/ε) matrix-vector products, which improves on the ubiquitous Hutchinson's estimator.

Fixed-Rank Approximation of a Positive-Semidefinite Matrix from Streaming Data

A new algorithm for fixed-rank psd matrix approximation from a sketch that combines the Nystrom approximation with a novel mechanism for rank truncation and exploits the spectral decay of the input matrix.

Krylov-aware stochastic trace estimation

This work introduces an algorithm for estimating the trace of a matrix function f ( A ) using implicit products with a symmetric matrix A and describes a Krylov subspace method for computing a low-rank approximation of a Matrix function by a computationally efficient projection onto Kryov subspace.

Monte Carlo Methods for Estimating the Diagonal of a Real Symmetric Matrix

The novel use of matrix concentration inequalities in the authors' proofs represents a systematic model for future analyses and implies that the accuracy of the estimators increases with the diagonal dominance of the matrix.

An estimator for the diagonal of a matrix

Fast Estimation of tr(f(A)) via Stochastic Lanczos Quadrature

An inexpensive method to estimate the trace of f(A) for cases where f is analytic inside a closed interval and A is a symmetric positive definite matrix, which combines three key ingredients, namely, the stochastic trace estimator, Gaussian quadrature, and the Lanczos algorithm.

Fast Randomized Kernel Ridge Regression with Statistical Guarantees

A version of this approach that comes with running time guarantees as well as improved guarantees on its statistical performance is described, and a fast algorithm is presented to quickly compute coarse approximations to these scores in time linear in the number of samples.