Linear programming boosting by column and row generation

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

We propose a new boosting algorithm based on a linear programming formulation. Our algorithm can take advantage of the sparsity of the solution of the underlying optimization problem. In preliminary experiments, our algorithm outperforms a state-of-the-art LP solver and LPBoost especially when the solution is given by a small set of relevant hypotheses and support vectors.

Original languageEnglish
Title of host publicationDiscovery Science - 12th International Conference, DS 2009, Proceedings
Pages401-408
Number of pages8
DOIs
Publication statusPublished - Nov 16 2009
Event12th International Conference on Discovery Science, DS 2009 - Porto, Portugal
Duration: Oct 3 2009Oct 5 2009

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume5808 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other12th International Conference on Discovery Science, DS 2009
CountryPortugal
CityPorto
Period10/3/0910/5/09

Fingerprint

Boosting
Linear programming
Support Vector
Sparsity
Optimization Problem
Formulation
Experiment
Experiments

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Hatano, K., & Takimoto, E. (2009). Linear programming boosting by column and row generation. In Discovery Science - 12th International Conference, DS 2009, Proceedings (pp. 401-408). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5808 LNAI). https://doi.org/10.1007/978-3-642-04747-3_32

Linear programming boosting by column and row generation. / Hatano, Kohei; Takimoto, Eiji.

Discovery Science - 12th International Conference, DS 2009, Proceedings. 2009. p. 401-408 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5808 LNAI).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Hatano, K & Takimoto, E 2009, Linear programming boosting by column and row generation. in Discovery Science - 12th International Conference, DS 2009, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 5808 LNAI, pp. 401-408, 12th International Conference on Discovery Science, DS 2009, Porto, Portugal, 10/3/09. https://doi.org/10.1007/978-3-642-04747-3_32
Hatano K, Takimoto E. Linear programming boosting by column and row generation. In Discovery Science - 12th International Conference, DS 2009, Proceedings. 2009. p. 401-408. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-04747-3_32
Hatano, Kohei ; Takimoto, Eiji. / Linear programming boosting by column and row generation. Discovery Science - 12th International Conference, DS 2009, Proceedings. 2009. pp. 401-408 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{63cb1b1c60814ee09893903f219f4aaa,
title = "Linear programming boosting by column and row generation",
abstract = "We propose a new boosting algorithm based on a linear programming formulation. Our algorithm can take advantage of the sparsity of the solution of the underlying optimization problem. In preliminary experiments, our algorithm outperforms a state-of-the-art LP solver and LPBoost especially when the solution is given by a small set of relevant hypotheses and support vectors.",
author = "Kohei Hatano and Eiji Takimoto",
year = "2009",
month = "11",
day = "16",
doi = "10.1007/978-3-642-04747-3_32",
language = "English",
isbn = "3642047467",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "401--408",
booktitle = "Discovery Science - 12th International Conference, DS 2009, Proceedings",

}

TY - GEN

T1 - Linear programming boosting by column and row generation

AU - Hatano, Kohei

AU - Takimoto, Eiji

PY - 2009/11/16

Y1 - 2009/11/16

N2 - We propose a new boosting algorithm based on a linear programming formulation. Our algorithm can take advantage of the sparsity of the solution of the underlying optimization problem. In preliminary experiments, our algorithm outperforms a state-of-the-art LP solver and LPBoost especially when the solution is given by a small set of relevant hypotheses and support vectors.

AB - We propose a new boosting algorithm based on a linear programming formulation. Our algorithm can take advantage of the sparsity of the solution of the underlying optimization problem. In preliminary experiments, our algorithm outperforms a state-of-the-art LP solver and LPBoost especially when the solution is given by a small set of relevant hypotheses and support vectors.

UR - http://www.scopus.com/inward/record.url?scp=71049179485&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=71049179485&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-04747-3_32

DO - 10.1007/978-3-642-04747-3_32

M3 - Conference contribution

AN - SCOPUS:71049179485

SN - 3642047467

SN - 9783642047466

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 401

EP - 408

BT - Discovery Science - 12th International Conference, DS 2009, Proceedings

ER -