gradient tree boosting implementation

1 次查看（过去 30 天）

Mahmoud Zeydabadinezhad 2017-10-18

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/361862-gradient-tree-boosting-implementation

评论： denny 2018-4-4

I need to implement gradient boosting with shrinkage in MATLAB. I'm allowed to use the built-in function(s) for decision tree. In gradient boosting (https://en.wikipedia.org/wiki/Gradient_boosting) there is a model update step which updates the current model as the sum of previously learned models and a weighted base learner (e.g. tree). My question is that if I use the fitrtree() function to train a base learner in each iteration, then how I should add the trained tree to update the model? The output of fitrtree() is a Regression Tree object and I don't know how to add two Regression Tree objects.

Thank you,