Expeditious Saliency-guided Mix-up through Random Gradient Thresholding

Open paper
Benchmark Model Rank Results
weakly-supervised-object-localization-on-imagenetR-Mix (ResNet-50)#6Top-1 Localization Accuracy: 55.58
image-classification-on-cifar-100R-Mix (WideResNet 28-10)#97Percentage correct: 85
image-classification-on-cifar-100RL-Mix (WideResNet 28-10)#99Percentage correct: 84.9
image-classification-on-cifar-100WideResNet 28-10 + CutMix (OneCycleLR scheduler)#107Percentage correct: 83.97
image-classification-on-cifar-100R-Mix (ResNeXt 29-4-24)#120Percentage correct: 83.02
image-classification-on-cifar-100RL-Mix (ResNeXt 29-4-24)#132Percentage correct: 82.43
image-classification-on-cifar-100R-Mix (WideResNet 16-8)#134Percentage correct: 82.32
image-classification-on-cifar-100ResNeXt 29-4-24 + CutMix (OneCycleLR scheduler)#135Percentage correct: 82.3
image-classification-on-cifar-100RL-Mix (WideResNet 16-8)#138Percentage correct: 82.16
image-classification-on-cifar-100WideResNet 16-8 + CutMix (OneCycleLR scheduler)#142Percentage correct: 81.79
image-classification-on-cifar-100R-Mix (PreActResNet-18)#146Percentage correct: 81.49
image-classification-on-cifar-100RL-Mix (PreActResNet-18)#154Percentage correct: 80.75
image-classification-on-cifar-100PreActResNet-18 + CutMix (OneCycleLR scheduler)#155Percentage correct: 80.6
image-classification-on-imagenetR-Mix (ResNet-50)#920Top 1 Accuracy: 77.39%

Rank counts only results with a code link.