Wavelet Decomposition of Gradient Boosting
Machine Learning
2019-05-06 v2 Signal Processing
Machine Learning
Abstract
In this paper we introduce a significant improvement to the popular tree-based Stochastic Gradient Boosting algorithm using a wavelet decomposition of the trees. This approach is based on harmonic analysis and approximation theoretical elements, and as we show through extensive experimentation, our wavelet based method generally outperforms existing methods, particularly in difficult scenarios of class unbalance and mislabeling in the training data.
Keywords
Cite
@article{arxiv.1805.02642,
title = {Wavelet Decomposition of Gradient Boosting},
author = {Shai Dekel and Oren Elisha and Ohad Morgan},
journal= {arXiv preprint arXiv:1805.02642},
year = {2019}
}