A study of tree-based methods and their combination
This work addresses efficiency improvements for tree-based methods, which are widely used in machine learning, but it appears incremental as it builds on existing frameworks.
The paper tackles the challenge of accelerating tree-based method fitting by introducing a general framework called importance sampled learning ensemble (ISLE) and proposes a model combination strategy, adaptive regression by mixing (ARM), along with three modified ISLEs, with performance evaluated on real datasets.
Tree-based methods are popular machine learning techniques used in various fields. In this work, we review their foundations and a general framework the importance sampled learning ensemble (ISLE) that accelerates their fitting process. Furthermore, we describe a model combination strategy called the adaptive regression by mixing (ARM), which is feasible for tree-based methods via ISLE. Moreover, three modified ISLEs are proposed, and their performance are evaluated on the real data sets.