Proceedings chapter
OA Policy
English

Multi-class Classification based on Binary Classifiers : On Coding Matrix Design, Reliability and Maximum Number of Classes

Presented at

Grenoble (France), 1-4 Sept. 2009

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Publication date

2009

Abstract

In this paper, we consider the multiclass classification problem based on independent set of binary classifiers. Each binary classifier represents the output of quantized projection of training data onto a randomly generated orthonormal basis vector thus producing a binary label. The ensemble of all binary labels forms an analogue of a coding matrix. The properties of such kind of matrices and their impact on the maximum number of uniquely distinguishable classes are analyzed in this paper from an information theoretic point of view. We also consider a concept of reliability for such kind of coding matrix generation that can be an alternative way for other adaptive training techniques and investigate the impact on the bit error probability. We demonstrate that it is equivalent to the considered random coding matrix without any bit reliability information in terms of recognition rate.

Keywords

  • Watermarking
  • error statistics
  • reliability
  • signal classification

Citation (ISO format)

VOLOSHYNOVSKYY, Svyatoslav et al. Multi-class Classification based on Binary Classifiers : On Coding Matrix Design, Reliability and Maximum Number of Classes. In: IEEE International Workshop on Machine Learning for Signal Processing, 2009, MLSP 2009. Grenoble (France). [s.l.] : Institute of Electrical and Electronics Engineers (IEEE), 2009. p. 1–6. doi: 10.1109/MLSP.2009.5306207

Main files (1)

Proceedings chapter (Published version)

Identifiers

616views
366downloads

Technical informations

Creation

06/03/2015 17:12:05

First validation

06/03/2015 17:12:05

Update

14/03/2023 22:58:25

Status update

14/03/2023 22:58:25

Last indexation

30/10/2024 23:16:52
All rights reserved by Archive ouverte UNIGE and the University of Geneva
OpenAIRE Indexed & Validated