Exploring Self-Supervised Regularization for Supervised and Semi-Supervised Learning

Benchmark Model Rank Results
semi-supervised-image-classification-on-cifar-10-1000-labelsSESEMI SSL (ConvNet)#7Accuracy: 82.12
semi-supervised-image-classification-on-svhn-250-labelsSESEMI SSL (ConvNet)#12Accuracy: 91.68
semi-supervised-image-classification-on-svhn-1000-labelsSESEMI SSL (ConvNet)#16Accuracy: 94.41
semi-supervised-image-classification-on-cifar-100-10000-labelsSESEMI SSL (ConvNet)#20Percentage error: 38.7
semi-supervised-image-classification-on-cifar-10-4000-labelsSESEMI SSL (ConvNet)#34Percentage error: 11.65