Machine Unlearning on Heart
0.033Unlearning Runtime (s)EUPG (k=30)
Evaluation results
| Method | Links | ||
|---|---|---|---|
| EUPG (k=30)Model Family=XGBoost, Forget ratio=5%, Number of runs=52025.07 | 0.033 | 2.74 | |
| EUPG (ϵ=1, δ=0)Model Family=XGBoost, Forget ratio=5%, Number of runs=52025.07 | 0.045 | 2.46 | |
| RetrainModel Family=XGBoost, Forget ratio=5%, Number of runs=52025.07 | 0.088 | 1 | |
| SISAModel Family=XGBoost, Forget ratio=5%, Number of runs=52025.07 | 0.228 | 0.33 | |
| EUPG (ϵ=1, δ=0)Model Family=Neural Network, Forget ratio=5%, Number of runs=52025.07 | 1.8 | 14.84 | |
| EUPG (k=30)Model Family=Neural Network, Forget ratio=5%, Number of runs=52025.07 | 1.82 | 14.76 | |
| Certified-SP (ϵ=1, δ=1e-5)Model Family=Neural Network, Forget ratio=5%, Number of runs=52025.07 | 1.84 | 14.04 | |
| RetrainModel Family=Neural Network, Forget ratio=5%, Number of runs=52025.07 | 31.75 | 1 | |
| SISAModel Family=Neural Network, Forget ratio=5%, Number of runs=52025.07 | 52.57 | 0.62 |