Image Classification on mini-ImageNet-Red 40% noise
65.6AccuracyMixup
Evaluation results
| Method | Links | |
|---|---|---|
| MixupBackbone=ResNet-18, Training Protocol=Oracle: Clean Subset, Data Augmentation=Mixup2022.02 | 65.6 | |
| StandardBackbone=ResNet-18, Training Protocol=Oracle: Clean Subset2022.02 | 64.8 | |
| NCRBackbone=ResNet-18, Training Protocol=NCR2022.02 | 64.6 | |
| NCR + MixupBackbone=ResNet-18, Training Protocol=NCR, Data Augmentation=Mixup2022.02 | 64.5 | |
| Label smoothingBackbone=ResNet-18, Training Protocol=Baseline2022.02 | 64.2 | |
| MixupBackbone=ResNet-18, Training Protocol=Baseline, Data Augmentation=Mixup2022.02 | 63.8 | |
| Label smoothing + MixupBackbone=ResNet-18, Training Protocol=Baseline, Data Augmentation=Mixup2022.02 | 63.8 | |
| BootstrapBackbone=ResNet-18, Training Protocol=Baseline2022.02 | 63.4 | |
| StandardBackbone=ResNet-18, Training Protocol=Baseline2022.02 | 63 | |
| Bootstrap + MixupBackbone=ResNet-18, Training Protocol=Baseline, Data Augmentation=Mixup2022.02 | 62 | |
| MOITBackbone=ResNet-18, Training Protocol=Other Works2022.02 | 60.8 | |
| ELRBackbone=ResNet-18, Training Protocol=Other Works2022.02 | 58.1 | |
| InstanceGM-SSself-supervised learning=true2022.09 | 56.37 | |
| PropMixself-supervised learning=true2022.09 | 56.22 | |
| InstanceGMself-supervised learning=false2022.09 | 52.24 | |
| D-MixBackbone=ResNet-18, Training Protocol=Other Works2022.02 | 50.3 | |
| FaMUSself-supervised learning=false2022.09 | 48.06 | |
| MentorMixself-supervised learning=false2022.09 | 47.14 | |
| DivideMixself-supervised learning=false2022.09 | 46.72 | |
| MixUpself-supervised learning=false2022.09 | 46.4 | |
| CEself-supervised learning=false2022.09 | 42.7 |