Ensemble Knowledge Distillation for Learning Improved and Efficient Networks

Open paper
Benchmark Model Rank Results
knowledge-distillation-on-imagenetADLIK-MO-P25(T:SeNet154, ResNet152b S:ResNet-50-prune25%)#17Top-1 accuracy %: 78.79model size: 56.9MCRD training setting: ✘
knowledge-distillation-on-imagenetADLIK-MO-P375(T:SeNet154, ResNet152b S:ResNet-50-prune37.5)#18Top-1 accuracy %: 78.07model size: 40.5MCRD training setting: ✘
knowledge-distillation-on-imagenetADLIK-MO-P50(T:SeNet154, ResNet152b S:ResNet-50-half)#24Top-1 accuracy %: 76.376model size: 27MCRD training setting: ✘

Rank counts only results with a code link.