DoubleMatch: Improving Semi-Supervised Learning with Self-Supervision

Benchmark Model Rank Results
semi-supervised-image-classification-on-cifarDoubleMatch#11Percentage error: 4.65±0.17
semi-supervised-image-classification-on-cifar-2DoubleMatch#3Percentage error: 21.22± 0.17
semi-supervised-image-classification-on-cifar-6DoubleMatch#10Percentage error: 5.56±0.42
semi-supervised-image-classification-on-cifar-7DoubleMatch#14Percentage error: 13.59±5.60
semi-supervised-image-classification-on-cifar-8DoubleMatch#11Percentage error: 41.83± 1.22
semi-supervised-image-classification-on-cifar-9DoubleMatch#8Percentage error: 27.07± 0.26
semi-supervised-image-classification-on-stl-1DoubleMatch#1Accuracy: 95.65±0.20
semi-supervised-image-classification-on-svhnDoubleMatch#2Accuracy: 97.90 ± 0.07
semi-supervised-image-classification-on-svhn-1DoubleMatch#3Accuracy: 97.63±0.35
semi-supervised-image-classification-on-svhn-2DoubleMatch#5Percentage error: 15.37±11.81