Stochastic natural gradient descent draws posterior samples in function space.

Samuel L. Smith Daniel Duckworth Quoc V. Le Jascha Sohl-Dickstein

Published in: CoRR (2018)

Keyphrases

error function
data sets
cost function
search space
state space
training samples
real world
support vector
artificial neural networks
probability distribution
loss function
vector space
posterior probability
learning rules