Boosting Discriminative Visual Representation Learning with Scenario-Agnostic Mixup

Open paper
Benchmark Model Rank Results
image-classification-on-places205SAMix (ResNet-50 Supervised)#9Top 1 Accuracy: 64.3
image-classification-on-tiny-imagenet-classificationResNeXt-50 (SAMix)#12Validation Acc: 72.18%
image-classification-on-tiny-imagenet-classificationResNet18 (SAMix)#15Validation Acc: 68.89%
image-classification-on-inaturalist-2018ResNeXt-101 (SAMix)#33Top-1 Accuracy: 70.54%
image-classification-on-inaturalist-2018ResNet-50 (SAMix)#41Top-1 Accuracy: 64.84%
image-classification-on-cifar-100WRN-28-8 +SAMix#87Percentage correct: 85.50
image-classification-on-cifar-100ResNeXt-50(32x4d) + SAMix#102Percentage correct: 84.42
image-classification-on-imagenetResNet-101 (SAMix)#710Top 1 Accuracy: 81.08%Number of params: 44.6M
image-classification-on-imagenetResNet-50 (SAMix)#791Top 1 Accuracy: 79.41%Number of params: 25.6M
image-classification-on-imagenetResNet-34 (SAMix)#956Top 1 Accuracy: 76.35%Number of params: 21.8M
image-classification-on-imagenetResNet-18 (SAMix)#1049Top 1 Accuracy: 72.33%Number of params: 11.7M

Rank counts only results with a code link.