Rule-enhanced penalized regression by column generation using rectangular maximum agreement

Jonathan Eckstein; Noam Goldberg; Ai Kagawa

Rule-enhanced penalized regression by column generation using rectangular maximum agreement

Jonathan Eckstein, Noam Goldberg, Ai Kagawa

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Scopus citations

Abstract

We describe a procedure enhancing Lx-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 AAP-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 method is computation-intensive, but has promising prediction performance.

Original language	English
Title of host publication	34th International Conference on Machine Learning, ICML 2017
Publisher	International Machine Learning Society (IMLS)
Pages	1762-1770
Number of pages	9
ISBN (Electronic)	9781510855144
State	Published - 1 Jan 2017
Externally published	Yes
Event	34th International Conference on Machine Learning, ICML 2017 - Sydney, Australia Duration: 6 Aug 2017 → 11 Aug 2017

Publication series

Name	34th International Conference on Machine Learning, ICML 2017
Volume	3

Conference

Conference	34th International Conference on Machine Learning, ICML 2017
Country/Territory	Australia
City	Sydney
Period	6/08/17 → 11/08/17

ASJC Scopus subject areas

Computational Theory and Mathematics
Human-Computer Interaction
Software

Cite this

@inproceedings{a176e6a2165b440e9ce418bafac6499e,

title = "Rule-enhanced penalized regression by column generation using rectangular maximum agreement",

abstract = "We describe a procedure enhancing Lx-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 AAP-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 method is computation-intensive, but has promising prediction performance.",

author = "Jonathan Eckstein and Noam Goldberg and Ai Kagawa",

year = "2017",

month = jan,

day = "1",

language = "English",

series = "34th International Conference on Machine Learning, ICML 2017",

publisher = "International Machine Learning Society (IMLS)",

pages = "1762--1770",

booktitle = "34th International Conference on Machine Learning, ICML 2017",

}

Eckstein, J, Goldberg, N & Kagawa, A 2017, Rule-enhanced penalized regression by column generation using rectangular maximum agreement. in 34th International Conference on Machine Learning, ICML 2017. 34th International Conference on Machine Learning, ICML 2017, vol. 3, International Machine Learning Society (IMLS), pp. 1762-1770, 34th International Conference on Machine Learning, ICML 2017, Sydney, Australia, 6/08/17.

Rule-enhanced penalized regression by column generation using rectangular maximum agreement. / Eckstein, Jonathan; Goldberg, Noam; Kagawa, Ai.
34th International Conference on Machine Learning, ICML 2017. International Machine Learning Society (IMLS), 2017. p. 1762-1770 (34th International Conference on Machine Learning, ICML 2017; Vol. 3).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Rule-enhanced penalized regression by column generation using rectangular maximum agreement

AU - Eckstein, Jonathan

AU - Goldberg, Noam

AU - Kagawa, Ai

PY - 2017/1/1

Y1 - 2017/1/1

N2 - We describe a procedure enhancing Lx-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 AAP-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 method is computation-intensive, but has promising prediction performance.

AB - We describe a procedure enhancing Lx-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 AAP-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 method is computation-intensive, but has promising prediction performance.

UR - http://www.scopus.com/inward/record.url?scp=85048435992&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85048435992

T3 - 34th International Conference on Machine Learning, ICML 2017

SP - 1762

EP - 1770

BT - 34th International Conference on Machine Learning, ICML 2017

PB - International Machine Learning Society (IMLS)

T2 - 34th International Conference on Machine Learning, ICML 2017

Y2 - 6 August 2017 through 11 August 2017

ER -

Rule-enhanced penalized regression by column generation using rectangular maximum agreement

Abstract

Publication series

Conference

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this