Fine-grained Image Classification on Aircraft (test)
84.17Best AccuracyDivideMix + SNSCL
Evaluation results
| Method | Links | ||
|---|---|---|---|
| DivideMix + SNSCLNoise Type=Asymmetric, Noise Ratio=10%2023.03 | 84.17 | 84.03 | |
| MLC + SNSCLNoise Type=Asymmetric, Noise Ratio=10%2023.03 | 82.75 | 82.43 | |
| SYM + SNSCLNoise Type=Asymmetric, Noise Ratio=10%2023.03 | 82.3 | 81.46 | |
| MW-Net + SNSCLNoise Type=Asymmetric, Noise Ratio=10%2023.03 | 80.69 | 80.22 | |
| JoCoR + SNSCLNoise Type=Asymmetric, Noise Ratio=10%2023.03 | 78.84 | 78.29 | |
| Label Smooth + SNSCLNoise Type=Asymmetric, Noise Ratio=10%2023.03 | 78.67 | 77.98 | |
| Conf. Penalty + SNSCLNoise Type=Asymmetric, Noise Ratio=10%2023.03 | 78.34 | 78.01 | |
| Cross-Entropy + SNSCLNoise Type=Asymmetric, Noise Ratio=10%2023.03 | 78.28 | 78.22 | |
| DivideMixNoise Type=Asymmetric, Noise Ratio=10%2023.03 | 76.57 | 76.24 | |
| DivideMix + SNSCLNoise Type=Asymmetric, Noise Ratio=30%2023.03 | 74.8 | 74.57 | |
| Co-teaching + SNSCLNoise Type=Asymmetric, Noise Ratio=10%2023.03 | 74.73 | 74.28 | |
| MLCNoise Type=Asymmetric, Noise Ratio=10%2023.03 | 74.4 | 73.91 | |
| GCE + SNSCLNoise Type=Asymmetric, Noise Ratio=10%2023.03 | 73.85 | 73.89 | |
| MW-NetNoise Type=Asymmetric, Noise Ratio=10%2023.03 | 72.76 | 70.18 | |
| MLC + SNSCLNoise Type=Asymmetric, Noise Ratio=30%2023.03 | 72.3 | 71.96 | |
| MW-Net + SNSCLNoise Type=Asymmetric, Noise Ratio=30%2023.03 | 70.49 | 69.9 | |
| SYM + SNSCLNoise Type=Asymmetric, Noise Ratio=30%2023.03 | 69.61 | 69.37 | |
| SYMNoise Type=Asymmetric, Noise Ratio=10%2023.03 | 69.31 | 67.45 | |
| JoCoR + SNSCLNoise Type=Asymmetric, Noise Ratio=30%2023.03 | 67.5 | 66.48 | |
| JoCoRNoise Type=Asymmetric, Noise Ratio=10%2023.03 | 67.12 | 64.99 | |
| Conf. Penalty + SNSCLNoise Type=Asymmetric, Noise Ratio=30%2023.03 | 66.88 | 66.34 | |
| Label Smooth + SNSCLNoise Type=Asymmetric, Noise Ratio=30%2023.03 | 66.28 | 65.56 | |
| Cross-EntropyNoise Type=Asymmetric, Noise Ratio=10%2023.03 | 65.98 | 62.53 | |
| Cross-Entropy + SNSCLNoise Type=Asymmetric, Noise Ratio=30%2023.03 | 65.44 | 65.11 | |
| Label SmoothNoise Type=Asymmetric, Noise Ratio=10%2023.03 | 65.29 | 63.34 | |
| Conf. PenaltyNoise Type=Asymmetric, Noise Ratio=10%2023.03 | 64.9 | 63.01 | |
| GCE + SNSCLNoise Type=Asymmetric, Noise Ratio=30%2023.03 | 64.33 | 63.91 | |
| DivideMixNoise Type=Asymmetric, Noise Ratio=30%2023.03 | 63.97 | 63.28 | |
| Co-teaching + SNSCLNoise Type=Asymmetric, Noise Ratio=30%2023.03 | 62.17 | 61.88 | |
| MLCNoise Type=Asymmetric, Noise Ratio=30%2023.03 | 59.44 | 59 | |
| Co-teachingNoise Type=Asymmetric, Noise Ratio=10%2023.03 | 55.94 | 49.85 | |
| MW-NetNoise Type=Asymmetric, Noise Ratio=30%2023.03 | 54.88 | 51.8 | |
| JoCoRNoise Type=Asymmetric, Noise Ratio=30%2023.03 | 52.25 | 50.28 | |
| Label SmoothNoise Type=Asymmetric, Noise Ratio=30%2023.03 | 51.88 | 47.71 | |
| Cross-EntropyNoise Type=Asymmetric, Noise Ratio=30%2023.03 | 51.1 | 47.85 | |
| SYMNoise Type=Asymmetric, Noise Ratio=30%2023.03 | 50.23 | 47.55 | |
| Conf. PenaltyNoise Type=Asymmetric, Noise Ratio=30%2023.03 | 49.38 | 47.53 | |
| Co-teachingNoise Type=Asymmetric, Noise Ratio=30%2023.03 | 45.18 | 38.97 | |
| GCENoise Type=Asymmetric, Noise Ratio=10%2023.03 | 44.22 | 44.16 | |
| GCENoise Type=Asymmetric, Noise Ratio=30%2023.03 | 34.18 | 33.66 |