Randomized Numerical Linear Algebra (RandNLA) is a powerful class of methods, widely used in High Performance Computing (HPC). RandNLA provides approximate solutions to linear algebra functions applied to large signals, at reduced computational costs. However, the randomization step for dimensionality reduction may itself become the computational bottleneck on traditional hardware. Leveraging near constant-time linear random projections delivered by LightOn Optical Processing Units we show that randomization can be significantly accelerated, at negligible precision loss, in a wide range of important RandNLA algorithms, such as RandSVD or trace estimators.
@article{arxiv.2104.14429,
title = {Photonic co-processors in HPC: using LightOn OPUs for Randomized Numerical Linear Algebra},
author = {Daniel Hesslow and Alessandro Cappelli and Igor Carron and Laurent Daudet and Raphaël Lafargue and Kilian Müller and Ruben Ohana and Gustave Pariente and Iacopo Poli},
journal= {arXiv preprint arXiv:2104.14429},
year = {2021}
}
Comments
Add "This project has received funding from the European Union's Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No 860830"