AI-READI
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Glucose forecasting | AI-READI Average across groups | iGlu-CE9.39 | 12 | |
| Glucose forecasting | AI-READI (Insulin T2D) | iGlu-CE13.04 | 12 | |
| Glucose forecasting | AI-READI (Non-Insulin T2D) | iGlu-CE10.31 | 12 | |
| Glucose forecasting | AI-READI (Pre-T2D) | iGlu-CE8.14 | 12 | |
| Glucose forecasting | AI-READI (Healthy) | iGlu-CE6.08 | 12 | |
| Glucose level forecasting (PH = 3 hours) | AI-READI Avg across groups | Mean Absolute Error (MAE)24.15 | 12 | |
| Glucose level forecasting (PH = 3 hours) | AI-READI Insulin T2D group | MAE33.32 | 12 | |
| Glucose level forecasting (PH = 3 hours) | AI-READI Non-Insulin T2D group | MAE23.62 | 12 | |
| Glucose level forecasting (PH = 3 hours) | AI-READI Pre-T2D group | MAE20.05 | 12 | |
| Glucose level forecasting (PH = 3 hours) | AI-READI Healthy group | MAE19.5 | 12 | |
| Diabetes Categorization | AI-READI (five random seeds) | Macro AUROC78.15 | 11 | |
| Counterfactual Generation | AI-READI (Class 1) | Validity98 | 9 | |
| Counterfactual Generation | AI-READI Class 0 | Validity0.99 | 9 | |
| Classification | AI-READI Scenario C — Dual-Class Undersampling (train) | ACC21.47 | 6 | |
| Classification | AI-READI Scenario B — Negative-Class Undersampling (train) | Accuracy17.16 | 6 | |
| Classification | AI-READI Scenario A — Positive-Class Undersampling (train) | ACC21 | 6 | |
| Membership Inference Attack | AI-READI | AUC100 | 4 | |
| Fidelity Evaluation | AI-READI (test) | CWC1.352 | 3 | |
| Classification | AI-READI Full (train) | Accuracy71.8 | 1 |
Showing 19 of 19 rows