Adaptive multi-module approximation network

Wonil Kim, Kishan Mehrotra, Chilukuri K. Mohan

Research output: Contribution to conferencePaper

1 Scopus citations

Abstract

The Adaptive Multi-module Approximation Network (AMAN) associates multiple reference vectors with each module in the network; network output depends on the output of the module whose reference vector is nearest an input vector. A competitive learning mechanism associates each module with reference vectors. New modules are introduced when performance of existing modules is inadequate for some training data.

Original languageEnglish (US)
Pages17-22
Number of pages6
StatePublished - Dec 1 1997
EventProceedings of the 1997 Artificial Neural Networks in Engineering Conference, ANNIE'97 - St.Louis, MO, USA
Duration: Nov 9 1997Nov 12 1997

Other

OtherProceedings of the 1997 Artificial Neural Networks in Engineering Conference, ANNIE'97
CitySt.Louis, MO, USA
Period11/9/9711/12/97

ASJC Scopus subject areas

  • Software

Cite this

Kim, W., Mehrotra, K., & Mohan, C. K. (1997). Adaptive multi-module approximation network. 17-22. Paper presented at Proceedings of the 1997 Artificial Neural Networks in Engineering Conference, ANNIE'97, St.Louis, MO, USA, .