Gaussian Compression Stream: Principle and Preliminary Results
Signal Processing
2020-11-13 v2 Information Theory
Machine Learning
math.IT
Abstract
Random projections became popular tools to process big data. In particular, when applied to Nonnegative Matrix Factorization (NMF), it was shown that structured random projections were far more efficient than classical strategies based on Gaussian compression. However, they remain costly and might not fully benefit from recent fast random projection techniques. In this paper, we thus investigate an alternative to structured ran-om projections-named Gaussian compression stream-which (i) is based on Gaussian compressions only, (ii) can benefit from the above fast techniques, and (iii) is shown to be well-suited to NMF.
Cite
@article{arxiv.2011.05390,
title = {Gaussian Compression Stream: Principle and Preliminary Results},
author = {Farouk Yahaya and Matthieu Puigt and Gilles Delmaire and Gilles Roussel},
journal= {arXiv preprint arXiv:2011.05390},
year = {2020}
}
Comments
in Proceedings of iTWIST'20, Paper-ID: 11, Nantes, France. December 2-4, 2020