Purchase
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Model Poisoning Attack | Purchase cross-device (test) | Iθ72.82 | 74 | |
| Federated Learning Model Poisoning Robustness | Purchase Cross-silo 100 FL clients, 500 global iterations & 3 layer DNN model | Attack Impact (I_theta)2.16 | 26 | |
| Membership Inference Attack | Purchase-100 | Accuracy97.91 | 8 | |
| Membership Inference Attack | Purchase-10 | Accuracy96.79 | 8 | |
| Membership Inference Attack | Purchase-50 | Accuracy96.77 | 6 | |
| Membership Inference Attack | Purchase-20 | Accuracy95.89 | 6 | |
| Membership Inference | Purchase-4 | Δ AUC0 | 4 | |
| Membership Inference | Purchase-3 | ΔAUC0 | 4 | |
| Membership Inference Attack | Purchase | Attack Accuracy85 | 4 | |
| Model Extraction | Purchase (DN) | Test Accuracy65.1 | 4 | |
| Federated Learning Classification | Purchase FedAvg 0.922 (test) | Accuracy92.6 | 4 | |
| Membership Inference Attack | Purchase-100 | PC62 | 3 | |
| Membership Inference Attack | Purchase100 (test) | Accuracy0.72 | 3 | |
| Model Poisoning Attack Control | Purchase100 (test) | Avg Control Error ξ15 | 2 |
Showing 14 of 14 rows