Scalable Feature Selection for (Multitask) Gradient Boosted Trees
This addresses the need for faster and more efficient feature selection in GBDT models used in search and recommendation systems, though it is incremental as it builds on existing methods.
The paper tackled the problem of inefficient feature selection in Gradient Boosted Decision Trees (GBDTs) for high-dimensional settings, developing a scalable forward feature selection method with group testing that achieved significant speedups in training time while maintaining competitive model performance.
Gradient Boosted Decision Trees (GBDTs) are widely used for building ranking and relevance models in search and recommendation. Considerations such as latency and interpretability dictate the use of as few features as possible to train these models. Feature selection in GBDT models typically involves heuristically ranking the features by importance and selecting the top few, or by performing a full backward feature elimination routine. On-the-fly feature selection methods proposed previously scale suboptimally with the number of features, which can be daunting in high dimensional settings. We develop a scalable forward feature selection variant for GBDT, via a novel group testing procedure that works well in high dimensions, and enjoys favorable theoretical performance and computational guarantees. We show via extensive experiments on both public and proprietary datasets that the proposed method offers significant speedups in training time, while being as competitive as existing GBDT methods in terms of model performance metrics. We also extend the method to the multitask setting, allowing the practitioner to select common features across tasks, as well as selecting task-specific features.