Classification on Synthetic corrupted data
100ASPTSpAM
Evaluation results
| Method | Links | ||
|---|---|---|---|
| TSpAMr1 (ratio of samples with noisy labels)=10%, r2 (imbalance factor)=50%2026.04 | 100 | 89 | |
| MAMr1 (ratio of samples with noisy labels)=10%, r2 (imbalance factor)=50%2026.04 | 100 | 91 | |
| SAMr1 (ratio of samples with noisy labels)=10%, r2 (imbalance factor)=50%2026.04 | 98 | 76 | |
| SpAMr1 (ratio of samples with noisy labels)=10%, r2 (imbalance factor)=50%2026.04 | 94 | 79 | |
| ℓ1-SVMr1 (ratio of samples with noisy labels)=30%, r2 (imbalance factor)=50%2026.04 | 53 | 57 | |
| TSpAMr1 (ratio of samples with noisy labels)=30%, r2 (imbalance factor)=50%2026.04 | 53 | 70 | |
| MAMr1 (ratio of samples with noisy labels)=30%, r2 (imbalance factor)=50%2026.04 | 53 | 72 | |
| ℓ1-SVMr1 (ratio of samples with noisy labels)=10%, r2 (imbalance factor)=50%2026.04 | 43 | 58 | |
| SpAMr1 (ratio of samples with noisy labels)=30%, r2 (imbalance factor)=50%2026.04 | 41 | 63 | |
| SAMr1 (ratio of samples with noisy labels)=30%, r2 (imbalance factor)=50%2026.04 | 38 | 60 | |
| MAMr1 (ratio of samples with noisy labels)=50%, r2 (imbalance factor)=50%2026.04 | 31 | 53 | |
| TSpAMr1 (ratio of samples with noisy labels)=50%, r2 (imbalance factor)=50%2026.04 | 27 | 50 | |
| ℓ1-SVMr1 (ratio of samples with noisy labels)=50%, r2 (imbalance factor)=50%2026.04 | 23 | 53 | |
| SAMr1 (ratio of samples with noisy labels)=50%, r2 (imbalance factor)=50%2026.04 | 3 | 50 | |
| SpAMr1 (ratio of samples with noisy labels)=50%, r2 (imbalance factor)=50%2026.04 | 1 | 50 |