Yale CS490 -- Special Projects -- Spring 1999
Project 2

Experiments on an Implementation for Randomized Singular Value Decomposition and Matrix Low-Rank Approximation

AUTHOR: Daniel Grossman (with Petros Drineas - Graduate Advisor & Ravindran Kannan - Faculty Advisor)
Abstract

Keywords:
Randomized Algorithms, Singular Value Decomposition, Low-Rank Approximation

Download my report in PostScript format here: Experiments on SVD/LRA

The report does not contain the MATLAB code discussed therein. Please access those four MATLAB modules here:
[exper10a.m] [mkmatrix.m] [qmult.m] [fastsvd.m]

Also find here an Excel workspace showing my raw data results, aggregate results, and a graphical explanation.

References

  1. A. Frieze, R. Kannan, and S. Vempala. "Fast Monte-Carlo Algorithms for Finding Low-Rank Approximations." Pre-print made available by author. October 22, 1998.
  2. P. Drineas, A. Frieze, R. Kanna, S. Vempala, and V. Vinay. "Clustering in Large Graphs and Matrices." Proceedings of the Symposium on Discrete Algorithms. SIAM, 1999.
  3. Golub, Gene H., and Charles F. Van Loan. Matrix Computations, Second Edition. The Johns Hopkins University Press, Baltimore, 1989.
This page last updated: Monday, October 11, 2004.
E-mail the author: Daniel Grossman