Gaussian Compression Stream: Principle and Preliminary Results

TL;DR

This paper introduces Gaussian compression stream, a new random projection method that leverages fast techniques for efficient processing of big data, especially for Nonnegative Matrix Factorization, with promising preliminary results.

Contribution

It proposes Gaussian compression stream, an alternative to structured random projections, combining Gaussian compressions with fast techniques for improved efficiency in NMF applications.

Findings

Efficient Gaussian compression stream suitable for NMF

Can leverage recent fast random projection techniques

Preliminary results show promising performance

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.