Image Classification on CIFAR-100 (Specific Robustness Attacks)
78.8Logit Flipping RobustnessMean
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| MeanHeterogeneity (α)=0.5, Clients (n)=17, Adversaries (f)=42025.09 | 78.8 | 72.8 | 66.6 | 66.4 | 56 | 39.2 | 39.3 | |
| DeepSet-TMHeterogeneity (α)=0.5, Clients (n)=17, Adversaries (f)=42025.09 | 78 | 74.7 | 76 | 76.4 | 63.7 | 49.6 | 49.6 | |
| GMHeterogeneity (α)=0.5, Clients (n)=17, Adversaries (f)=42025.09 | 75.4 | 71.5 | 75.4 | 75.3 | 59.6 | 39 | 39 | |
| CWTMHeterogeneity (α)=0.5, Clients (n)=17, Adversaries (f)=42025.09 | 74.8 | 71.5 | 74.9 | 74.8 | 60.8 | 44.9 | 44.9 | |
| CWMedHeterogeneity (α)=0.5, Clients (n)=17, Adversaries (f)=42025.09 | 65.8 | 62.3 | 66.1 | 66.1 | 62.9 | 41.7 | 41.7 |