Masks, Signs, And Learning Rate Rewinding.

Advait Harshal Gadhikar Rebekka Burkholz

Published in: ICLR (2024)

Keyphrases

learning rate
convergence rate
learning algorithm
hidden layer
error function
convergence speed
adaptive learning rate
multilayer neural networks
training algorithm
edge detection
rapid convergence
weight vector
activation function
primal dual