Why does an MVS gradient-sum estimate weight sampled examples by inverse inclusion probability?
Classic MLLevel: AdvancedCatBoost · Bootstrap
Why does an MVS gradient-sum estimate weight sampled examples by inverse inclusion probability?
Hints
- Sampling is deliberately unequal.
- A ratio of unbiased estimates needs separate analysis.
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: CatBoost · Bootstrap · Apache-2.0 · answer: Quantoros editorial