An Empirical Study on Regularization of Deep Neural Networks by Local Rademacher Complexity.

Yingzhen Yang Xingjian Li Jun Huan

Published in: CoRR (2019)

Keyphrases

rademacher complexity
data dependent
risk bounds
error bounds
manifold regularization
generalization error
semi supervised learning
reproducing kernel hilbert space
empirical risk minimization
generalization bounds
training set
semi supervised
learning rate
sample complexity
vc dimension