Using Pre-Training Can Improve Model Robustness and Uncertainty.

Dan Hendrycks Kimin Lee Mantas Mazeika

Published in: ICML (2019)

Keyphrases

mathematical model
statistical model
formal model
high level
probability distribution
partial information
training samples
parameter estimation
computational model
experimental data
stochastic nature
machine learning
decision theory
training phase
em algorithm
input data
e learning