IPBoost -- Non-Convex Boosting via Integer Programming
Machine Learning
2020-02-13 v1 Optimization and Control
Machine Learning
Abstract
Recently non-convex optimization approaches for solving machine learning problems have gained significant attention. In this paper we explore non-convex boosting in classification by means of integer programming and demonstrate real-world practicability of the approach while circumventing shortcomings of convex boosting approaches. We report results that are comparable to or better than the current state-of-the-art.
Cite
@article{arxiv.2002.04679,
title = {IPBoost -- Non-Convex Boosting via Integer Programming},
author = {Marc E. Pfetsch and Sebastian Pokutta},
journal= {arXiv preprint arXiv:2002.04679},
year = {2020}
}