New algorithm for self-organizing neural classifiers suitable for easy hardware implementation

Slawomir Przylucki; Konrad Plachecki; Mariusz Duk

doi:10.1117/12.532389

24 October 2003 New algorithm for self-organizing neural classifiers suitable for easy hardware implementation

Slawomir Przylucki, Konrad Plachecki, Mariusz Duk

Proceedings Volume 5125, Photonics Applications in Astronomy, Communications, Industry, and High-Energy Physics Experiments; (2003) https://doi.org/10.1117/12.532389
Event: Photonics Applications in Astronomy, Communications, Industry, and High-Energy Physics Experiments 2002, 2002, Wilga, Poland

Abstract

Artificial neural networks (ANN), or connectionist classifiers, are massively parallel computation systems that are based on simplified models of the human brain. Their complex classifications capabilities, combined with properties such as generalization, fault-tolerance and learning make them attractive for a range of applications that conventional computers found difficult. One of the possible neural net applications is an analysis of high dimension data sets. Thanks to mentioned above classifications capabilities, net output signals are low-dimension representations of inputs where each output can represented some input signal feature. In this paper we present the new algorihtm of multivariate data classification. The algorithm based on modified counterpropagation neural network. The main goal of our research as to develop a new classifier architecture which reduces the required number of interconnection in a hidden layer as well as output layer. That allows easier hardware implementation of proposed algorithm.

Citation Download Citation

Slawomir Przylucki, Konrad Plachecki, and Mariusz Duk "New algorithm for self-organizing neural classifiers suitable for easy hardware implementation", Proc. SPIE 5125, Photonics Applications in Astronomy, Communications, Industry, and High-Energy Physics Experiments, (24 October 2003); https://doi.org/10.1117/12.532389

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available