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

13 Apr 2017Takeru MiyatoShin-ichi MaedaMasanori KoyamaShin Ishii

We propose a new regularization method based on virtual adversarial loss: a new measure of local smoothness of the conditional label distribution given input. Virtual adversarial loss is defined as the robustness of the conditional label distribution around each input data point against local perturbation... (read more)

PDF Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT LEADERBOARD
Semi-Supervised Image Classification cifar10, 250 Labels VAT Percentage correct 63.97 # 4
Semi-Supervised Image Classification CIFAR-10, 250 Labels VAT Accuracy 63.97 # 6
Semi-Supervised Image Classification CIFAR-10, 4000 Labels VAT Accuracy 88.64 # 13
Semi-Supervised Image Classification CIFAR-10, 4000 Labels VAT+EntMin Accuracy 89.45 # 12
Semi-Supervised Image Classification ImageNet - 10% labeled data VAT + EntMin (Miyato et al., 2018) Top 5 Accuracy Conv-Large # 26
Semi-Supervised Image Classification SVHN, 1000 labels VAT+EntMin Accuracy 96.14 # 6
Semi-Supervised Image Classification SVHN, 1000 labels VAT Accuracy 94.58 # 10
Semi-Supervised Image Classification SVHN, 250 Labels VAT Accuracy 91.59 # 7

Methods used in the Paper


METHOD TYPE
🤖 No Methods Found Help the community by adding them if they're not listed; e.g. Deep Residual Learning for Image Recognition uses ResNet