Skip to Main content Skip to Navigation
Conference papers

Gaussian Compression Stream: Principle and Preliminary Results

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 random 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.
Complete list of metadata

Cited literature [21 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02931454
Contributor : Matthieu Puigt <>
Submitted on : Thursday, November 12, 2020 - 5:27:09 PM
Last modification on : Tuesday, January 5, 2021 - 1:04:02 PM
Long-term archiving on: : Saturday, February 13, 2021 - 8:15:58 PM

File

fympgdgr_iTwist_2020.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02931454, version 1

Collections

Citation

Farouk Yahaya, Matthieu Puigt, Gilles Delmaire, Gilles Roussel. Gaussian Compression Stream: Principle and Preliminary Results. iTWIST : international Traveling Workshop on Interactions between low-complexity data models and Sensing Techniques, Dec 2020, Nantes, France. Paper-ID: 11. ⟨hal-02931454⟩

Share

Metrics

Record views

38

Files downloads

11