Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Document limitation of one-split-at-a-time Greedy tree learning heuristic (#4233)

Browse Source
This commit is contained in:
Philip Hyunsu Cho 2019-03-08 10:05:39 -08:00 committed by GitHub
parent 617f572c0f
commit 331cd3e4f7
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 4 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
doc/tutorials/model.rst
View File

@ -255,6 +255,10 @@ For real valued data, we usually want to search for an optimal split. To efficie
A left to right scan is sufficient to calculate the structure score of all possible split solutions, and we can find the best split efficiently.
.. note:: Limitation of additive tree learning
Since it is intractable to enumerate all possible tree structures, we add one split at a time. This approach works well most of the time, but there are some edge cases that fail due to this approach. For those edge cases, training results in a degenerate model because we consider only one feature dimension at a time. See `Can Gradient Boosting Learn Simple Arithmetic? <http://mariofilho.com/can-gradient-boosting-learn-simple-arithmetic/>`_ for an example.
**********************
Final words on XGBoost
**********************