Efficient coordinate-wise leading eigenvector computation
We develop and analyze efficient "coordinate-wise" methods for finding the leading eigenvector, where each step involves only a vector-vector product. We establish global convergence with overall runtime guarantees that are at least as good as Lanczos's method and dominate it for slow...
Saved in:
Main Authors | , , , |
---|---|
Format | Journal Article |
Language | English |
Published |
25.02.2017
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | We develop and analyze efficient "coordinate-wise" methods for finding the
leading eigenvector, where each step involves only a vector-vector product. We
establish global convergence with overall runtime guarantees that are at least
as good as Lanczos's method and dominate it for slowly decaying spectrum. Our
methods are based on combining a shift-and-invert approach with coordinate-wise
algorithms for linear regression. |
---|---|
DOI: | 10.48550/arxiv.1702.07834 |