Boosting the Performance of Semi-Supervised Learning with Unsupervised Clustering

Benchmark Model Rank Results
semi-supervised-image-classification-on-stl-10-1000-labelsSemi-MMDC#2Accuracy: 95.22±0.29
semi-supervised-image-classification-on-svhn-250-labelsSemi-MMDC#2Accuracy: 97.7±0.03
semi-supervised-image-classification-on-svhn-40-labelsSemi-MMDC#2Percentage error: 3.09±0.54
semi-supervised-image-classification-on-cifar-10-10-labelsSemi-MMDC#3Accuracy (Test): 70.84±8.1
semi-supervised-image-classification-on-cifar-10-20-labelsSemi-MMDC#3Percentage error: 28.1±5.5
semi-supervised-image-classification-on-cifar-10-250-labelsSemi-MMDC#9Percentage error: 5.51±0.25
semi-supervised-image-classification-on-cifar-10-40-labelsSemi-MMDC#12Percentage error: 7.39±0.61