Using feature transformation and selection with polynomial networks

W. M. Campbell, Huan Liu

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

Abstract

Polynomial networks have proven successful in authentication applications such as speaker recognition. A drawback of these methods is that as the degree of the polynomial network is increased, the number of model terms increases rapidly. This rapid increase can result in overfitting and make the network difficult to use in real-world applications because of the large number of model terms. We propose and contrast two solutions to this problem. First, we show how random dimension reduction can be used to effectively control model complexity. We describe a novel method which allows quick reduction of the dimension using an FFT. Applying these methods to a speaker recognition problem shows an approximately linear relation between the log of the number of model parameters and the log of the error rate. Second, we apply several methods of feature selection to reduce both model complexity and computation. We survey several methods and show which method yields the best performance in a speaker recognition application.

Original languageEnglish (US)
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
EditorsK.L. Priddy, P.E. Keller, P.J. Angeline
Pages175-186
Number of pages12
Volume4390
DOIs
StatePublished - 2001
EventApplications and Science of Computational Intelligence IV - Orlando, FL, United States
Duration: Apr 17 2001Apr 18 2001

Other

OtherApplications and Science of Computational Intelligence IV
Country/TerritoryUnited States
CityOrlando, FL
Period4/17/014/18/01

Keywords

  • Feature selection
  • Feature transformation
  • Pattern classification
  • Polynomials
  • Speaker recognition
  • User authentication

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Fingerprint

Dive into the research topics of 'Using feature transformation and selection with polynomial networks'. Together they form a unique fingerprint.

Cite this