Contrastive Regularization for Semi-Supervised Learning

Benchmark Model Rank Results
semi-supervised-image-classification-on-cifar-10-250-labelsFixMatch+CR–Percentage error: 5.04
semi-supervised-image-classification-on-cifar-10-40-labelsFixMatch+CR–Percentage error: 5.69
semi-supervised-image-classification-on-cifar-10-4000-labelsFixMatch+CR–Percentage error: 4.16
semi-supervised-image-classification-on-cifar-100-10000-labelsFixMatch+CR–Percentage error: 21.03
semi-supervised-image-classification-on-cifar-100-2500-labelsFixMatch+CR–Percentage error: 27.58
semi-supervised-image-classification-on-cifar-100-400-labelsFixMatch+CR–Percentage error: 49.23