RMSProp and equilibrated adaptive learning rates for non-convex optimization.

Yann N. Dauphin Harm de Vries Junyoung Chung Yoshua Bengio

Published in: CoRR (2015)

Keyphrases

convex optimization
learning rate
low rank
covering numbers
primal dual
learning algorithm
convergence rate
convex optimization problems
total variation
gaussian kernels
uniform convergence
augmented lagrangian
training data
upper bound