A Sparse SVD Method for High-dimensional Data

Details A Sparse SVD Method for High-dimensional Data A Sparse SVD Method for High-dimensional Data

2011-12-12

arxiv.org/abs/1112.2433v1

Statistics

Methodology

Scientific paper

We present a new computational approach to approximating a large, noisy data table by a low-rank matrix with sparse singular vectors. The approximation is obtained from thresholded subspace iterations that produce the singular vectors simultaneously, rather than successively as in competing proposals. We introduce novel ways to estimate thresholding parameters which obviate the need for computationally expensive cross-validation. We also introduce a way to sparsely initialize the algorithm for computational savings that allow our algorithm to outperform the vanilla SVD on the full data table when the signal is sparse. A comparison with two existing sparse SVD methods suggests that our algorithm is computationally always faster and statistically always at least comparable to the better of the two competing algorithms.

Affiliated with

Also associated with

No associations

Rating

A Sparse SVD Method for High-dimensional Data does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with A Sparse SVD Method for High-dimensional Data, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and A Sparse SVD Method for High-dimensional Data will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-708329