Linear programming boosting by column and row generation

研究成果: 著書/レポートタイプへの貢献会議での発言

2 引用 (Scopus)

抄録

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.

元の言語英語
ホスト出版物のタイトルDiscovery Science - 12th International Conference, DS 2009, Proceedings
ページ401-408
ページ数8
DOI
出版物ステータス出版済み - 11 16 2009
イベント12th International Conference on Discovery Science, DS 2009 - Porto, ポルトガル
継続期間: 10 3 200910 5 2009

出版物シリーズ

名前Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
5808 LNAI
ISSN(印刷物)0302-9743
ISSN(電子版)1611-3349

その他

その他12th International Conference on Discovery Science, DS 2009
ポルトガル
Porto
期間10/3/0910/5/09

Fingerprint

Boosting
Linear programming
Support Vector
Sparsity
Optimization Problem
Formulation
Experiment
Experiments

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

これを引用

Hatano, K., & Takimoto, E. (2009). Linear programming boosting by column and row generation. : 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); 巻数 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); 巻 5808 LNAI).

研究成果: 著書/レポートタイプへの貢献会議での発言

Hatano, K & Takimoto, E 2009, Linear programming boosting by column and row generation. : 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), 巻. 5808 LNAI, pp. 401-408, 12th International Conference on Discovery Science, DS 2009, Porto, ポルトガル, 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. : 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 -