Smooth boosting for margin-based ranking

Jun Ichi Moribe; Kohei Hatano; Eiji Takimoto; Masayuki Takeda

doi:10.1007/978-3-540-87987-9_21

Smooth boosting for margin-based ranking

Jun Ichi Moribe, Kohei Hatano, Eiji Takimoto, Masayuki Takeda

Research output: Contribution to journal › Conference article › peer-review

3 Citations (Scopus)

Abstract

We propose a new boosting algorithm for bipartite ranking problems. Our boosting algorithm, called SoftRankBoost, is a modification of RankBoost which maintains only smooth distributions over data. SoftRankBoost provably achieves approximately the maximum soft margin over all pairs of positive and negative examples, which implies high AUC score for future data.

Original language	English
Pages (from-to)	227-239
Number of pages	13
Journal	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	5254 LNAI
DOIs	https://doi.org/10.1007/978-3-540-87987-9_21
Publication status	Published - 2008
Event	19th International Conference on Algorithmic Learning Theory, ALT 2008 - Budapest, Hungary Duration: Oct 13 2008 → Oct 16 2008

All Science Journal Classification (ASJC) codes

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-540-87987-9_21

Cite this

@article{18f55b20f5bd4bdc9b955e191648d0ca,

title = "Smooth boosting for margin-based ranking",

abstract = "We propose a new boosting algorithm for bipartite ranking problems. Our boosting algorithm, called SoftRankBoost, is a modification of RankBoost which maintains only smooth distributions over data. SoftRankBoost provably achieves approximately the maximum soft margin over all pairs of positive and negative examples, which implies high AUC score for future data.",

author = "Moribe, {Jun Ichi} and Kohei Hatano and Eiji Takimoto and Masayuki Takeda",

year = "2008",

doi = "10.1007/978-3-540-87987-9_21",

language = "English",

volume = "5254 LNAI",

pages = "227--239",

journal = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

issn = "0302-9743",

publisher = "Springer Science and Business Media Deutschland GmbH",

note = "19th International Conference on Algorithmic Learning Theory, ALT 2008 ; Conference date: 13-10-2008 Through 16-10-2008",

}

TY - JOUR

T1 - Smooth boosting for margin-based ranking

AU - Moribe, Jun Ichi

AU - Hatano, Kohei

AU - Takimoto, Eiji

AU - Takeda, Masayuki

PY - 2008

Y1 - 2008

N2 - We propose a new boosting algorithm for bipartite ranking problems. Our boosting algorithm, called SoftRankBoost, is a modification of RankBoost which maintains only smooth distributions over data. SoftRankBoost provably achieves approximately the maximum soft margin over all pairs of positive and negative examples, which implies high AUC score for future data.

AB - We propose a new boosting algorithm for bipartite ranking problems. Our boosting algorithm, called SoftRankBoost, is a modification of RankBoost which maintains only smooth distributions over data. SoftRankBoost provably achieves approximately the maximum soft margin over all pairs of positive and negative examples, which implies high AUC score for future data.

UR - http://www.scopus.com/inward/record.url?scp=56749160429&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=56749160429&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-87987-9_21

DO - 10.1007/978-3-540-87987-9_21

M3 - Conference article

AN - SCOPUS:56749160429

SN - 0302-9743

VL - 5254 LNAI

SP - 227

EP - 239

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

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

T2 - 19th International Conference on Algorithmic Learning Theory, ALT 2008

Y2 - 13 October 2008 through 16 October 2008

ER -

Smooth boosting for margin-based ranking

Abstract

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this