Image Classification on mini-ImageNet-Blue 40% noise
67.3AccuracyMixup
Evaluation results
| Method | Links | |
|---|---|---|
| MixupBackbone=ResNet-18, Training Protocol=Oracle: Clean Subset, Data Augmentation=Mixup2022.02 | 67.3 | |
| StandardBackbone=ResNet-18, Training Protocol=Oracle: Clean Subset2022.02 | 67.2 | |
| NCR + MixupBackbone=ResNet-18, Training Protocol=NCR, Data Augmentation=Mixup2022.02 | 61.4 | |
| Label smoothing + MixupBackbone=ResNet-18, Training Protocol=Baseline, Data Augmentation=Mixup2022.02 | 60.9 | |
| Label smoothingBackbone=ResNet-18, Training Protocol=Baseline2022.02 | 60.6 | |
| NCRBackbone=ResNet-18, Training Protocol=NCR2022.02 | 60.6 | |
| MixupBackbone=ResNet-18, Training Protocol=Baseline, Data Augmentation=Mixup2022.02 | 59.4 | |
| BootstrapBackbone=ResNet-18, Training Protocol=Baseline2022.02 | 57.4 | |
| StandardBackbone=ResNet-18, Training Protocol=Baseline2022.02 | 55.9 | |
| Bootstrap + MixupBackbone=ResNet-18, Training Protocol=Baseline, Data Augmentation=Mixup2022.02 | 53.2 |