IPBoost – Non-Convex Boosting via Integer Programming

Marc Pfetsch, Sebastian Pokutta
Proceedings of the 37th International Conference on Machine Learning, PMLR 119:7663-7672, 2020.

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 circumvent- ing shortcomings of convex boosting approaches. We report results that are comparable to or better than the current state-of-the-art.

Cite this Paper


BibTeX
@InProceedings{pmlr-v119-pfetsch20a, title = {{IPB}oost {–} Non-Convex Boosting via Integer Programming}, author = {Pfetsch, Marc and Pokutta, Sebastian}, booktitle = {Proceedings of the 37th International Conference on Machine Learning}, pages = {7663--7672}, year = {2020}, editor = {III, Hal Daumé and Singh, Aarti}, volume = {119}, series = {Proceedings of Machine Learning Research}, month = {13--18 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v119/pfetsch20a/pfetsch20a.pdf}, url = {http://proceedings.mlr.press/v119/pfetsch20a.html}, 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 circumvent- ing shortcomings of convex boosting approaches. We report results that are comparable to or better than the current state-of-the-art.} }
Endnote
%0 Conference Paper %T IPBoost – Non-Convex Boosting via Integer Programming %A Marc Pfetsch %A Sebastian Pokutta %B Proceedings of the 37th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2020 %E Hal Daumé III %E Aarti Singh %F pmlr-v119-pfetsch20a %I PMLR %P 7663--7672 %U http://proceedings.mlr.press/v119/pfetsch20a.html %V 119 %X 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 circumvent- ing shortcomings of convex boosting approaches. We report results that are comparable to or better than the current state-of-the-art.
APA
Pfetsch, M. & Pokutta, S.. (2020). IPBoost – Non-Convex Boosting via Integer Programming. Proceedings of the 37th International Conference on Machine Learning, in Proceedings of Machine Learning Research 119:7663-7672 Available from http://proceedings.mlr.press/v119/pfetsch20a.html.

Related Material