Image Classification on CIFAR-10-LT λ=10 (test)
9.94Error RateHybrid-PSC
Evaluation results
| Method | Links | |
|---|---|---|
| Hybrid-PSCBackbone=ResNet-322021.12 | 9.94 | |
| MetaSAug with cross-entropy lossBackbone=ResNet-322021.12 | 10.56 | |
| RISDABackbone=ResNet-322021.12 | 10.64 | |
| Meta-class-weight with cross-entropy lossBackbone=ResNet-322021.12 | 11.15 | |
| BBNBackbone=ResNet-322021.12 | 11.68 | |
| Meta-weight netBackbone=ResNet-322021.12 | 12.45 | |
| MixupBackbone=ResNet-322021.12 | 12.9 | |
| Class-balanced CEBackbone=ResNet-322021.12 | 13.1 | |
| Cross-EntropyBackbone=ResNet-322021.12 | 13.82 | |
| Class-balanced fine-tuningBackbone=ResNet-322021.12 | 16.83 | |
| L2RWBackbone=ResNet-322021.12 | 17.88 |