Approximate reduction from AUC maximization to 1-norm soft margin optimization

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

2 Citations (Scopus)

Abstract

Finding linear classifiers that maximize AUC scores is important in ranking research. This is naturally formulated as a 1-norm hard/soft margin optimization problem over pn pairs of p positive and n negative instances. However, directly solving the optimization problems is impractical since the problem size (pn) is quadratically larger than the given sample size (p+n). In this paper, we give (approximate) reductions from the problems to hard/soft margin optimization problems of linear size. First, for the hard margin case, we show that the problem is reduced to a hard margin optimization problem over p+n instances in which the bias constant term is to be optimized. Then, for the soft margin case, we show that the problem is approximately reduced to a soft margin optimization problem over p+n instances for which the resulting linear classifier is guaranteed to have a certain margin over pairs.

Original languageEnglish
Title of host publicationAlgorithmic Learning Theory - 22nd International Conference, ALT 2011, Proceedings
Pages324-337
Number of pages14
DOIs
Publication statusPublished - Oct 20 2011
Event22nd International Conference on Algorithmic Learning Theory, ALT 2011 - Espoo, Finland
Duration: Oct 5 2011Oct 7 2011

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume6925 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other22nd International Conference on Algorithmic Learning Theory, ALT 2011
CountryFinland
CityEspoo
Period10/5/1110/7/11

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'Approximate reduction from AUC maximization to 1-norm soft margin optimization'. Together they form a unique fingerprint.

  • Cite this

    Suehiro, D., hatano, K., & Takimoto, E. (2011). Approximate reduction from AUC maximization to 1-norm soft margin optimization. In Algorithmic Learning Theory - 22nd International Conference, ALT 2011, Proceedings (pp. 324-337). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6925 LNAI). https://doi.org/10.1007/978-3-642-24412-4_26