Faster SGD training by minibatch persistency.

Matteo Fischetti Iacopo Mandatelli Domenico Salvagnin

Published in: CoRR (2018)

Keyphrases

stochastic gradient descent
training process
training set
training speed
computer software
training phase
supervised learning
online learning
training samples
multi layer perceptron
training algorithm
step size
logistic regression
real time
feature vectors
case study
social networks
machine learning
data mining