ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoring

Benchmark Model Rank Results
semi-supervised-image-classification-on-stl-10-1000-labelsReMixMatch#5Accuracy: 93.82
semi-supervised-image-classification-on-svhn-1000-labelsReMixMatch#6Accuracy: 97.17
semi-supervised-image-classification-on-cifar-100-2500-labelsReMixMatch#9Percentage error: 27.43±0.31
semi-supervised-image-classification-on-cifar-10-250-labelsReMixMatch#11Percentage error: 6.27
semi-supervised-image-classification-on-cifar-100-400-labelsReMixMatch#13Percentage error: 44.28±2.06
semi-supervised-image-classification-on-cifar-10-40-labelsReMixMatch#15Percentage error: 19.10
semi-supervised-image-classification-on-cifar-10-4000-labelsReMixMatch#15Percentage error: 5.14
image-classification-on-stl-10ReMixMatch (K=4)#17Percentage correct: 93.82
image-classification-on-stl-10ReMixMatch (K=1)#19Percentage correct: 93.23
image-classification-on-stl-10MixMatch#28Percentage correct: 89.82
image-classification-on-stl-10CC-GAN#48Percentage correct: 77.80
image-classification-on-stl-10SWWAE#55Percentage correct: 74.30