Weighted order statistic classifiers with large rank-order margin. Metadata

Metadata describes a digital item, providing (if known) such information as creator, publisher, contents, size, relationship to other resources, and more. Metadata may also contain "preservation" components that help us to maintain the integrity of digital files over time.

Title

  • Main Title Weighted order statistic classifiers with large rank-order margin.

Creator

  • Author: Porter, R. B. (Reid B.)
    Creator Type: Personal
  • Author: Hush, D. R. (Donald R.)
    Creator Type: Personal
  • Author: Theiler, J. P. (James P.)
    Creator Type: Personal
  • Author: Gokhale, M. (Maya)
    Creator Type: Personal

Contributor

  • Sponsor: United States. Department of Energy.
    Contributor Type: Organization

Publisher

  • Name: Los Alamos National Laboratory
    Place of Publication: United States

Date

  • Creation: 2003-01-01

Language

  • English

Description

  • Content Description: We describe how Stack Filters and Weighted Order Statistic function classes can be used for classification problems. This leads to a new design criteria for linear classifiers when inputs are binary-valued and weights are positive . We present a rank-based measure of margin that can be directly optimized as a standard linear program and investigate its effect on generalization error with experiment. Our approach can robustly combine large numbers of base hypothesis and easily implement known priors through regularization.
  • Physical Description: [9] p.

Subject

  • Keyword: Statistics
  • Keyword: Classification
  • Keyword: Computers
  • Keyword: Hypothesis
  • STI Subject Categories: 97 Mathematical Methods And Computing
  • Keyword: Learning
  • Keyword: Design

Source

  • Conference: Submitted to: 20th International Conference on Machine Learning, August 2003, Washington

Collection

  • Name: Office of Scientific & Technical Information Technical Reports
    Code: OSTI

Institution

  • Name: UNT Libraries Government Documents Department
    Code: UNTGD

Resource Type

  • Article

Format

  • Text

Identifier

  • Report No.: LA-UR-03-0545
  • Grant Number: none
  • Office of Scientific & Technical Information Report Number: 976524
  • Archival Resource Key: ark:/67531/metadc928443