Image Classification on CIFAR-10 (test) (Noise Robustness)
96.7Accuracy (20% Noise)L2B-C2D
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| L2B-C2DNoise type=Symmetric2022.02 | 96.7 | 95.6 | 94.8 | 94.4 | |
| L2B-UniConNoise type=Symmetric2022.02 | 96.5 | 95.8 | 94.7 | 92.8 | |
| AugDescNoise type=Symmetric2022.02 | 96.3 | 95.4 | 93.8 | 91.9 | |
| C2DNoise type=Symmetric2022.02 | 96.3 | 95.2 | 94.4 | 93.5 | |
| DivideMixNoise type=Symmetric2022.02 | 96.1 | 94.6 | 93.2 | 76 | |
| L2B-DivideMixNoise type=Symmetric2022.02 | 96.1 | 95.4 | 94 | 91.3 | |
| UniConNoise type=Symmetric2022.02 | 96 | 95.6 | 93.9 | 90.8 | |
| MixupNoise type=Symmetric2022.02 | 95.6 | 87.1 | 71.6 | 52.2 | |
| Sel-CL+Noise type=Symmetric2022.02 | 95.5 | 93.9 | 89.2 | 81.9 | |
| TCLNoise type=Symmetric2022.02 | 95 | 93.9 | 92.5 | 89.4 | |
| MOIT+Noise type=Symmetric2022.02 | 94.1 | 91.8 | 81.1 | 74.7 | |
| M-correctionNoise type=Symmetric2022.02 | 94 | 92 | 86.8 | 69.1 | |
| MSLCNoise type=Symmetric2022.02 | 93.4 | 89.9 | 69.8 | 56.1 | |
| Meta-LearningNoise type=Symmetric2022.02 | 92.9 | 89.3 | 77.4 | 58.7 | |
| MLCNoise type=Symmetric2022.02 | 92.6 | 88.1 | 77.4 | 67.9 | |
| PENCILNoise type=Symmetric2022.02 | 92.4 | 89.1 | 77.5 | 58.9 | |
| GCENoise type=Symmetric2022.02 | 90 | 89.3 | 73.9 | 36.5 | |
| Co-teaching+Noise type=Symmetric2022.02 | 89.5 | 85.7 | 67.4 | 47.9 |