Rule-Enhanced Penalized Regression by Column Generation using Rectangular Maximum Agreement

Jonathan Eckstein, Noam Goldberg, Ai Kagawa
Proceedings of the 34th International Conference on Machine Learning, PMLR 70:1059-1067, 2017.

Abstract

We describe a learning procedure enhancing L1-penalized regression by adding dynamically generated rules describing multidimensional “box” sets. Our rule-adding procedure is based on the classical column generation method for high-dimensional linear programming. The pricing problem for our column generation procedure reduces to the NP-hard rectangular maximum agreement (RMA) problem of finding a box that best discriminates between two weighted datasets. We solve this problem exactly using a parallel branch-and-bound procedure. The resulting rule-enhanced regression procedure is computation-intensive, but has promising prediction performance.

Cite this Paper

BibTeX
@InProceedings{pmlr-v70-eckstein17a, title = {Rule-Enhanced Penalized Regression by Column Generation using Rectangular Maximum Agreement}, author = {Jonathan Eckstein and Noam Goldberg and Ai Kagawa}, booktitle = {Proceedings of the 34th International Conference on Machine Learning}, pages = {1059--1067}, year = {2017}, editor = {Precup, Doina and Teh, Yee Whye}, volume = {70}, series = {Proceedings of Machine Learning Research}, month = {06--11 Aug}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v70/eckstein17a/eckstein17a.pdf}, url = {https://proceedings.mlr.press/v70/eckstein17a.html}, abstract = {We describe a learning procedure enhancing L1-penalized regression by adding dynamically generated rules describing multidimensional “box” sets. Our rule-adding procedure is based on the classical column generation method for high-dimensional linear programming. The pricing problem for our column generation procedure reduces to the NP-hard rectangular maximum agreement (RMA) problem of finding a box that best discriminates between two weighted datasets. We solve this problem exactly using a parallel branch-and-bound procedure. The resulting rule-enhanced regression procedure is computation-intensive, but has promising prediction performance.} }
Endnote
%0 Conference Paper %T Rule-Enhanced Penalized Regression by Column Generation using Rectangular Maximum Agreement %A Jonathan Eckstein %A Noam Goldberg %A Ai Kagawa %B Proceedings of the 34th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2017 %E Doina Precup %E Yee Whye Teh %F pmlr-v70-eckstein17a %I PMLR %P 1059--1067 %U https://proceedings.mlr.press/v70/eckstein17a.html %V 70 %X We describe a learning procedure enhancing L1-penalized regression by adding dynamically generated rules describing multidimensional “box” sets. Our rule-adding procedure is based on the classical column generation method for high-dimensional linear programming. The pricing problem for our column generation procedure reduces to the NP-hard rectangular maximum agreement (RMA) problem of finding a box that best discriminates between two weighted datasets. We solve this problem exactly using a parallel branch-and-bound procedure. The resulting rule-enhanced regression procedure is computation-intensive, but has promising prediction performance.
APA
Eckstein, J., Goldberg, N. & Kagawa, A.. (2017). Rule-Enhanced Penalized Regression by Column Generation using Rectangular Maximum Agreement. Proceedings of the 34th International Conference on Machine Learning, in Proceedings of Machine Learning Research 70:1059-1067 Available from https://proceedings.mlr.press/v70/eckstein17a.html.

Related Material