Feature Interactions in XGBoost
arXiv:2007.05758v112 citations
AI Analysis
This work addresses model performance and interpretability for users of XGBoost, but it appears incremental as it builds on existing methods.
The paper tackles the problem of identifying feature interactions to use as constraints in XGBoost gradient boosting tree models, resulting in significant performance improvements and better interpretability.
In this paper, we investigate how feature interactions can be identified to be used as constraints in the gradient boosting tree models using XGBoost's implementation. Our results show that accurate identification of these constraints can help improve the performance of baseline XGBoost model significantly. Further, the improvement in the model structure can also lead to better interpretability.