Optimizing for the decision not the prediction: an exploration of Smooth Net Benefit as a training objective
The results do not support $\sigma$NB as a general replacement for NLL training, but support further investigation of decision-focused objectives in settings where conventional likelihood-based training may not adequately capture decision-relevant structure.
Koen M. F. Gorgels, L. Barreñada, M. van Smeden et al.
· 0 citations