Distilling the Knowledge in a Neural Network

Open paper
Benchmark Model Rank Results
knowledge-distillation-on-cifar-100resnet8x4 (T: resnet32x4 S: resnet8x4)#21Top-1 Accuracy (%): 73.33
knowledge-distillation-on-cifar-100vgg8 (T:vgg13 S:vgg8)#22Top-1 Accuracy (%): 72.98
knowledge-distillation-on-cifar-100resnet110 (T:resnet110 S:resnet20)#27Top-1 Accuracy (%): 70.67
knowledge-distillation-on-imagenetADLIK-MO(T: ResNet101 S: ResNet50)#22Top-1 accuracy %: 77.14model size: 99MCRD training setting: ✘
knowledge-distillation-on-imagenetKD (T: ResNet-34 S:ResNet-18)#53Top-1 accuracy %: 70.66CRD training setting: ✓

Rank counts only results with a code link.