Image Classification on CIFAR-100 clean (test) (Epsilon-Dependent Accuracy)
72.91Accuracy (ε=0.2)CORES2*
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| CORES2*Noise type=Instance-based, Variant=Star2020.10 | 72.91 | 70.66 | 63.08 | |
| CORES2Noise type=Instance-based2020.10 | 61.25 | 47.81 | 37.85 | |
| Peer LossNoise type=Instance-based2020.10 | 59.92 | 45.76 | 33.61 | |
| SCENoise type=Instance-based2020.10 | 59.87 | 41.76 | 23.41 | |
| Forward TNoise type=Instance-based2020.10 | 58.95 | 41.68 | 22.83 | |
| Cross EntropyNoise type=Instance-based2020.10 | 58.72 | 41.14 | 25.29 | |
| LDMINoise type=Instance-based2020.10 | 58.66 | 41.77 | 28 | |
| LqNoise type=Instance-based2020.10 | 58.18 | 40.32 | 23.13 | |
| JoCoRNoise type=Instance-based2020.10 | 44.28 | 22.77 | 7.54 | |
| Co-teachingNoise type=Instance-based2020.10 | 43.03 | 23.13 | 7.07 | |
| Co-teaching+Noise type=Instance-based2020.10 | 41.84 | 24.4 | 8.74 |