This question was migrated from the Slack channel #questions.
Original Question:
Hello everyone, can someone point me to the paper (if it exists) that proposes the bagging aggregation strategy used for XGBoost?
Answer:
Hi! The bagging aggregation strategy used for XGBoost is not based on a specific paper, rather, it is a common and effective way to do tree ensemble, which can be used for federated XGBoost.Here are some related materials to help understand this method:
- Paper: https://arxiv.org/pdf/2011.02796.pdf, it’s about how to do FL with Gradient boosting decision tree (GBDT), XGBoost belongs to GBDT.
- Wikipedia of bootstrap bagging: Bootstrap aggregating - Wikipedia