Transfer Learning for Human & AI

We combine the design of two random number generators to obtain a new class of generators with heavy-weight characteristic polynomials . These generators passed all tests of the \emph{TestU01}-package for each 32-bit integer part and each64-bit derived real part of the output . We determine the size gaps for 32-bits words, neglecting the non-linear tempering, and compare with analternative experimental linear tempering .

Author(s) : Marcel Van de Vel

Links : PDF - Abstract

Code :


Keywords : generators - tempering - linear - part - bit -

• Machine Learning Nanodegree
• The Hundred-Page Machine Learning Book
• Python for Everybody on Coursera
• Neural Networks from Scratch with Python by Sentdex
• Deep Learning Nanodegree
• fast.ai Machine Learning
• Data Science A-Z
• AI for Everyone by Andrew Ng