Why can a loss coupling every pair of training rows not be dropped directly into a standard XGBoost custom objective?
Classic MLLevel: AdvancedXGBoost · Custom objectives
Why can a loss coupling every pair of training rows not be dropped directly into a standard XGBoost custom objective?
Hints
- What is the unit of a returned derivative?
- Cross-row Hessian blocks are not represented.
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