Image Classification on CIFAR-10 (Accuracy and Calibration Metrics)
96AccuracyKC
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| KCBackbone=ConvNeXt2026.05 | 96 | 0.0038 | 0.5501 | 0.0008 | 0.0025 | 0.122 | — | |
| LNBackbone=ConvNeXt2026.05 | 96 | 0.0032 | 0.5615 | 0.0007 | 0.0023 | 0.122 | — | |
| SMBackbone=ConvNeXt2026.05 | 95.9 | 0.0036 | 0.5791 | 0.0006 | 0.0021 | 0.124 | — | |
| VQBackbone=ConvNeXt2026.05 | 95.8 | 0.0024 | 0.3326 | 0.0006 | 0.0021 | 0.128 | — | |
| DCBackbone=ConvNeXt2026.05 | 95.8 | 0.0034 | 0.581 | 0.0006 | 0.002 | 0.124 | — | |
| IRBackbone=ConvNeXt2026.05 | 95.7 | 0.004 | 0.5589 | 0.0006 | 0.0024 | 0.148 | — | |
| PSBackbone=ConvNeXt2026.05 | 95.6 | 0.0077 | 0.7295 | 0.0013 | 0.0059 | 0.187 | — | |
| TSBackbone=ConvNeXt2026.05 | 95.6 | 0.0047 | 0.5893 | 0.0014 | 0.0033 | 0.135 | — | |
| NCBackbone=ConvNeXt2026.05 | 95.6 | 0.0043 | 0.64 | 0.0017 | 0.0039 | 0.143 | — | |
| PCBackbone=ConvNeXt2026.05 | 95.5 | 0.0045 | 0.5987 | 0.0014 | 0.0032 | 0.139 | — | |
| DEData Augmentation=Random crop and horizontal flip, Samples=102024.08 | 93.1 | — | — | — | 0.017 | 0.209 | 0.293 | |
| L2EData Augmentation=Random crop and horizontal flip, Samples=10, Temperature (T)=0.00012024.08 | 92.6 | — | — | — | 0.037 | 0.235 | 0.391 | |
| CSGMCMCData Augmentation=Random crop and horizontal flip, Samples=10, Temperature (T)=0.00012024.08 | 92.3 | — | — | — | 0.031 | 0.234 | 0.142 | |
| FedPerNumber of Clients=20, Clustering Strategy=Clustering (MMDSW)2026.06 | 84.2 | — | — | — | — | — | — | |
| FedRepNumber of Clients=20, Clustering Strategy=Clustering (MMDSW)2026.06 | 83.2 | — | — | — | — | — | — | |
| FedPerNumber of Clients=40, Clustering Strategy=Clustering (MMDSW)2026.06 | 82.9 | — | — | — | — | — | — | |
| FedPerNumber of Clients=20, Clustering Strategy=Vanilla2026.06 | 81.9 | — | — | — | — | — | — | |
| FedPerNumber of Clients=40, Clustering Strategy=Vanilla2026.06 | 81.7 | — | — | — | — | — | — | |
| FedRepNumber of Clients=40, Clustering Strategy=Clustering (MMDSW)2026.06 | 81.1 | — | — | — | — | — | — | |
| FedRepNumber of Clients=20, Clustering Strategy=Vanilla2026.06 | 80.7 | — | — | — | — | — | — | |
| FedPerNumber of Clients=100, Clustering Strategy=Clustering (MMDSW)2026.06 | 80.5 | — | — | — | — | — | — | |
| FedRepNumber of Clients=40, Clustering Strategy=Vanilla2026.06 | 78.2 | — | — | — | — | — | — | |
| FedPerNumber of Clients=100, Clustering Strategy=Vanilla2026.06 | 77.8 | — | — | — | — | — | — | |
| FedRepNumber of Clients=100, Clustering Strategy=Clustering (MMDSW)2026.06 | 72.9 | — | — | — | — | — | — | |
| FedRepNumber of Clients=100, Clustering Strategy=Vanilla2026.06 | 70.4 | — | — | — | — | — | — | |
| FedAvgNumber of Clients=20, Clustering Strategy=Clustering (MMDSW)2026.06 | 60.6 | — | — | — | — | — | — | |
| FedAvgNumber of Clients=40, Clustering Strategy=Clustering (MMDSW)2026.06 | 57.5 | — | — | — | — | — | — | |
| FedAvgNumber of Clients=100, Clustering Strategy=Clustering (MMDSW)2026.06 | 47.7 | — | — | — | — | — | — | |
| FedAvgNumber of Clients=20, Clustering Strategy=Vanilla2026.06 | 35.5 | — | — | — | — | — | — | |
| FedAvgNumber of Clients=40, Clustering Strategy=Vanilla2026.06 | 35.2 | — | — | — | — | — | — | |
| FedAvgNumber of Clients=100, Clustering Strategy=Vanilla2026.06 | 31.2 | — | — | — | — | — | — | |
| OTDDMetric Type=Average Uplift2026.06 | 18.8 | — | — | — | — | — | — | |
| MMDSWMetric Type=Average Uplift2026.06 | 15.7 | — | — | — | — | — | — |