Long-tailed Image Classification on CIFAR-100-LT λ=20 (test)
41.33Error RateRISDA
Evaluation results
| Method | Links | |
|---|---|---|
| RISDABackbone=ResNet-322021.12 | 41.33 | |
| MetaSAug with cross-entropy lossBackbone=ResNet-32, Reference (Source of result)=Original paper2021.12 | 42.15 | |
| Meta-class-weight with cross-entropy lossBackbone=ResNet-32, Reference (Source of result)=Li et al. 20212021.12 | 44.38 | |
| Meta-weight netBackbone=ResNet-32, Reference (Source of result)=Li et al. 20212021.12 | 46.96 | |
| Class-balanced fine-tuningBackbone=ResNet-32, Reference (Source of result)=Li et al. 20212021.12 | 47.7 | |
| Class-balanced CEBackbone=ResNet-32, Reference (Source of result)=Li et al. 20212021.12 | 48.06 | |
| Cross-EntropyBackbone=ResNet-322021.12 | 48.94 | |
| L2RWBackbone=ResNet-32, Reference (Source of result)=Li et al. 20212021.12 | 49.25 |