Embedded systems are utilizing complex machine learning designs to solve difficult problems. It is a challenge to maximize the design efficiency with limitations to space, power, and heat generation. Random Number Generators (RNGs) must meet design constraints while also trying to be sufficiently random. We investigate the randomness of multiple RNGs and explore how degrees of randomness affect machine learning.
Verification of Random Number Generators for Embedded Machine Learning
01.07.2018
833202 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Pseudo-Random Number Generators
NTRS | 1984
|FPGA based Hybrid Random Number Generators
IEEE | 2020
|On properties of random number generators and their influence on traffic simulation
Tema Archiv | 1983
|