Toward generalizable and interpretable machine learning models in healthcare: Insights from ICU outcome predictions.

In MIMIC-IV, the models identify meaningful relationships: LR shows a negative trend, whereas EBM and XGB capture a U-shaped curve linking very small values to higher mortality.

also in 2,482 other papers

Taken together, these consistent findings across all tasks highlight an important trend: as model complexity increases, so does the amount of local data needed for a purely local model to surpass one using additional external data.

also in 625 other papers