Classification by Components Including Chow’s Reject Option Konferenzpaper uri icon

 

Abstract

  • In this paper, we present an approach how reject options are integrated into Classification-by-Component networks. Classification-by-Component networks are relatively accurate classifiers that offer a fair interpretability. Yet, their performance can be increased by allowing rejection of an uncertain classification. We will modify the original Classification-by-Component model so that the adaptive parameters adapt, taking reject options into account.

Veröffentlichungszeitpunkt

  • 2023

Zugangsrechte

  • false

Startseite

  • 586

letzte Seite

  • 596

Seitenzahl

  • 10

Internationale Standardbuchnummer (ISBN) 13

  • 9789819916382
  • 9789819916399