CatBoost
Stub. Created as a placeholder — expand with vault-sourced content.
What it is
Yandex’s gradient-boosted decision tree library. Offers ranking objectives such as YetiRank, PairLogit, and QueryRMSE, making it a LambdaMART-family option alongside LightGBM and XGBoost. Strong native handling of categorical features.
Vault references
- Hybrid Search and Learning-to-Rank with Metarank — names XGBoost, LightGBM, and CatBoost as the major LambdaMART implementations, all handling missing feature values via decision-tree default split routes
TODO
- Verify objective names against current CatBoost docs.