Iteratively reweighted fitting for reduced multivariate polynomial model

Wangmeng Zuo, Kuanquan Wang, Dapeng Zhang, Feng Yue

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review


Recently a class of reduced multivariate polynomial models (RM) has been proposed that performs well in classification tasks involving few features and many training data. The RM method, however, adopts a ridge leastsquare estimator, overlooking the fact that least square error usually does not correspond to minimum classification error. In this paper, we propose an iteratively reweighted regression method and two novel weight functions for fitting the RM model (IRF-RM). The IRF-RM method iteratively increases the weights of samples prone to misclassification and decreases the weights of samples far from the decision boundary, making the IRF-RM model more suitable for efficient pattern classification. A number of benchmark data sets are used to evaluate the IRF-RM method. Experimental results indicate that IRF-RM achieves a higher or comparable classification accuracy compared with RM and several state-of-the-art classification approaches.
Original languageEnglish
Title of host publicationAdvances in Neural Networks - ISNN 2007 - 4th International Symposium on Neural Networks, ISNN 2007, Proceedings
Number of pages10
EditionPART 2
Publication statusPublished - 24 Dec 2007
Event4th International Symposium on Neural Networks, ISNN 2007 - Nanjing, China
Duration: 3 Jun 20077 Jun 2007

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 2
Volume4492 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference4th International Symposium on Neural Networks, ISNN 2007

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science


Dive into the research topics of 'Iteratively reweighted fitting for reduced multivariate polynomial model'. Together they form a unique fingerprint.

Cite this