Image Classification on CIFAR-100 Flip2 noise 1.0 (test)
65.37Top-1 AccMW-net
Evaluation Results
| Method | Links | |
|---|---|---|
| MW-netNoise level p=0.2, Backbone=ResNet-322022.11 | 65.37 | |
| CrossEntropyNoise level p=0.2, Backbone=ResNet-322022.11 | 64.91 | |
| Advisor NetworkNoise level p=0.2, Backbone=ResNet-322022.11 | 63.07 | |
| MW-netNoise level p=0.4, Backbone=ResNet-322022.11 | 59 | |
| Advisor NetworkNoise level p=0.4, Backbone=ResNet-322022.11 | 57.96 | |
| CrossEntropyNoise level p=0.4, Backbone=ResNet-322022.11 | 57.7 | |
| Advisor NetworkNoise level p=0.6, Backbone=ResNet-322022.11 | 45.35 | |
| MW-netNoise level p=0.6, Backbone=ResNet-322022.11 | 36.97 | |
| CrossEntropyNoise level p=0.6, Backbone=ResNet-322022.11 | 36.55 | |
| Advisor NetworkNoise level p=0.8, Backbone=ResNet-322022.11 | 22.41 | |
| MW-netNoise level p=0.8, Backbone=ResNet-322022.11 | 7.99 | |
| CrossEntropyNoise level p=0.8, Backbone=ResNet-322022.11 | 7 |