High dimensional model representation (HDMR) based folded vector decomposition

  • Authors:
  • Letİsya Dİvanyan;Metİn Demİralp

  • Affiliations:
  • İstanbul Technical University, Informatics Institute, İstanbul, Türkİye;İstanbul Technical University, Informatics Institute, İstanbul, Türkİye

  • Venue:
  • AICT'11 Proceedings of the 2nd international conference on Applied informatics and computing theory
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

The "folded vector" statement which has been accepted by many scientists recently can be employed for naming multiindex arrays some mathematicians prefer to call "tensors" despite their wrong recalling some physical features they do not have here in fact. Folded vectors can be considered as arrays having more than one indices. The purpose of this work is to decompose a folded vector by using High Dimensional Model Representation (HDMR) which has now a quite powerful theory behind it. To this end, the considered array is decomposed to a constant term followed by one index, two indices, three indices, and so on, components. Paper presents the basic conceptual issues to be used for this purpose together with some comments about the practical applications.