Login / Signup
Existence and Estimation of Critical Batch Size for Training Generative Adversarial Networks with Two Time-Scale Update Rule.
Naoki Sato
Hideaki Iiduka
Published in:
ICML (2023)
Keyphrases
</>
batch size
update rule
batch mode
stochastic gradient descent
generative model
training samples
incremental learning
fixpoint
single item
training set
ant colony optimization algorithm
learning rate
poisson process
probabilistic model
neural network
online learning
active learning
learning algorithm