Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning

Benchmark Model Rank Results
semi-supervised-image-classification-on-svhn-250-labelsVAT#13Accuracy: 91.59
semi-supervised-image-classification-on-cifar-10-250-labelsVAT#15Percentage error: 36.03
semi-supervised-image-classification-on-svhn-1000-labelsVAT#15Accuracy: 94.58
semi-supervised-image-classification-on-cifar-10-4000-labelsVAT+EntMin#31Percentage error: 10.55
semi-supervised-image-classification-on-cifar-10-4000-labelsVAT#33Percentage error: 11.36