Deep Stable neural networks: large-width asymptotics and convergence rates.

Stefano Favaro Sandra Fortini Stefano Peluchetti

Published in: CoRR (2021)

Keyphrases

convergence rate
neural network
convergence speed
learning rate
sufficient conditions
primal dual
global convergence
numerical stability
mutation operator
artificial neural networks
back propagation
gaussian kernels
hidden layer
markov chain
conjugate gradient
radial basis function
stopping criterion