What can go wrong if a custom XGBoost loss returns negative Hessian entries?
Classic MLLevel: AdvancedXGBoost · Custom objectives
What can go wrong if a custom XGBoost loss returns negative Hessian entries?
Hints
- The optimizer assumes useful convex curvature.
- Clipping changes the local model of the loss.
Answer
This answer is part of mentoring
Advanced questions come with a full breakdown and a mock interview with a working quant researcher. The first call is free.
Source and licence: XGBoost · Custom objectives · Apache-2.0 · answer: Quantoros editorial