Share Email Print
cover

Proceedings Paper

Generating good design from bad design: dynamical network approach
Author(s): Mohammed R. Sayeh; A. Ragu
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

This paper discusses a dynamical network for mapping of interciass members without performing a learning process. This allows a member of class A to be mapped to a member of class B. Given sample members of each class a backpropagation network is trained to form the corresponding class boundaries. Upon completion of the training process the weights obtained are used in a recurrent network which performs the interclass member mapping without any further training. This mapping is achieved as the recurrent network evolves In time. The Initial state of the network is mapped to its equilibrium state. The interclass member mapping network (IMMN) has many applications in selfcorrecting systems. In this paper the IMMN is developed to represent two classes namely class B (for instance a class for representing members with desirable and correct features) and class A (members with incorrect features). An example is given in which two categories are used namely poorly and well-designed manufacturing parts. Given a poorly-designed part the network wifi suggest corrections resulting in a well-designed part. This example has nonlinear decision regions and shows the generalization capability of the network.

Paper Details

Date Published: 1 March 1991
PDF: 9 pages
Proc. SPIE 1396, Applications of Optical Engineering: Proceedings of OE/Midwest '90, (1 March 1991); doi: 10.1117/12.47760
Show Author Affiliations
Mohammed R. Sayeh, Southern Illinois Univ./Carbondale (United States)
A. Ragu, Southern Illinois Univ./Carbondale (United States)


Published in SPIE Proceedings Vol. 1396:
Applications of Optical Engineering: Proceedings of OE/Midwest '90

© SPIE. Terms of Use
Back to Top