Why random reshuffling beats stochastic gradient descent.

Mert Gürbüzbalaban Asuman E. Ozdaglar Pablo A. Parrilo

Published in: Math. Program. (2021)

Keyphrases

stochastic gradient descent
least squares
loss function
step size
matrix factorization
random forests
multiple kernel learning
weight vector
online algorithms
regularization parameter
importance sampling
feature space
particle filter
learning problems