Feature Prioritization and Regularization Improve Standard Accuracy and Adversarial Robustness.

Chihuang Liu Joseph F. JáJá

Published in: IJCAI (2019)

Keyphrases

computational efficiency
prediction accuracy
improve the recognition accuracy
neural network
feature vectors
computational cost
classification accuracy
multi class
image classification
error rate
highly accurate
data sets
computer vision
information systems
multi agent
image features