Intelligent system for automatic feature detection and selection or identification
Abstract
A neural network uses a fuzzy membership function, the parameters of which are adaptive during the training process, to parameterize the interconnection weights between an (n-1)'th layer and an n'th layer of the network. Each j'th node in each k'th layer of the network except the input layer produces its output value y.sub.k,j according to the function ##EQU1## where N.sub.k-1 is the number of nodes in layer k-1, i indexes the nodes of layer k-1 and all the w.sub.k,i,j are interconnection weights. The interconnection weights to all nodes j in the n'th layer are given by w.sub.n,i,j =w.sub.n,j (i, p.sub.n,j,1, . . . , p.sub.n,j,p.sbsb.n). The apparatus is trained by setting values for at least one of the parameters p.sub.n,j,1, . . . , p.sub.n,j,Pn. Preferably the number of parameters P.sub.n is less than the number of nodes N.sub.n-1 in layer n-1. w.sub.n,j (i,p.sub.n,j,1, . . . , p.sub.n,j,Pn) can be convex in i, and it can be bell-shaped. Sample functions for w.sub.n,j (i, p.sub.n,j,1, . . . , p.sub.n,j,Pn) include ##EQU2##
- Inventors:
-
- Pao-Shan Shiang, TW
- Framingham, MA
- San Francisco, CA
- Issue Date:
- Research Org.:
- Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
- OSTI Identifier:
- 871136
- Patent Number(s):
- 5664066
- Assignee:
- United States of America as represented by United States (Washington, DC)
- Patent Classifications (CPCs):
-
H - ELECTRICITY H01 - BASIC ELECTRIC ELEMENTS H01J - ELECTRIC DISCHARGE TUBES OR DISCHARGE LAMPS
- DOE Contract Number:
- W-7405-ENG-48
- Resource Type:
- Patent
- Country of Publication:
- United States
- Language:
- English
- Subject:
- intelligent; automatic; feature; detection; selection; identification; neural; network; fuzzy; membership; function; parameters; adaptive; training; process; parameterize; interconnection; weights; n-1; layer; node; input; produces; output; value; according; equ1; k-1; nodes; indexes; sbsb; apparatus; trained; setting; values; pn; preferably; convex; bell-shaped; sample; functions; equ2; neural network; training process; feature detection; neural net; output value; /706/
Citation Formats
Sun, Chuen-Tsai, Jang, Jyh-Shing, and Fu, Chi-Yung. Intelligent system for automatic feature detection and selection or identification. United States: N. p., 1997.
Web.
Sun, Chuen-Tsai, Jang, Jyh-Shing, & Fu, Chi-Yung. Intelligent system for automatic feature detection and selection or identification. United States.
Sun, Chuen-Tsai, Jang, Jyh-Shing, and Fu, Chi-Yung. Wed .
"Intelligent system for automatic feature detection and selection or identification". United States. https://www.osti.gov/servlets/purl/871136.
@article{osti_871136,
title = {Intelligent system for automatic feature detection and selection or identification},
author = {Sun, Chuen-Tsai and Jang, Jyh-Shing and Fu, Chi-Yung},
abstractNote = {A neural network uses a fuzzy membership function, the parameters of which are adaptive during the training process, to parameterize the interconnection weights between an (n-1)'th layer and an n'th layer of the network. Each j'th node in each k'th layer of the network except the input layer produces its output value y.sub.k,j according to the function ##EQU1## where N.sub.k-1 is the number of nodes in layer k-1, i indexes the nodes of layer k-1 and all the w.sub.k,i,j are interconnection weights. The interconnection weights to all nodes j in the n'th layer are given by w.sub.n,i,j =w.sub.n,j (i, p.sub.n,j,1, . . . , p.sub.n,j,p.sbsb.n). The apparatus is trained by setting values for at least one of the parameters p.sub.n,j,1, . . . , p.sub.n,j,Pn. Preferably the number of parameters P.sub.n is less than the number of nodes N.sub.n-1 in layer n-1. w.sub.n,j (i,p.sub.n,j,1, . . . , p.sub.n,j,Pn) can be convex in i, and it can be bell-shaped. Sample functions for w.sub.n,j (i, p.sub.n,j,1, . . . , p.sub.n,j,Pn) include ##EQU2##},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Wed Jan 01 00:00:00 EST 1997},
month = {Wed Jan 01 00:00:00 EST 1997}
}
Works referenced in this record:
A method for self-determination of adaptive learning rates in back propagation
journal, January 1991
- Weir, Michael K.
- Neural Networks, Vol. 4, Issue 3
Neural-network-based fuzzy logic control and decision system
journal, January 1991
- Lin, C. -T.; Lee, C. S. G.
- IEEE Transactions on Computers, Vol. 40, Issue 12
Initializing multilayer neural networks with fuzzy logic
conference, January 1992
- Okada, H.; Watanabe, N.; Kawamura, A.
- [1992] IJCNN International Joint Conference on Neural Networks, [Proceedings 1992] IJCNN International Joint Conference on Neural Networks
An introduction to computing with neural nets
journal, January 1987
- Lippmann, R.
- IEEE ASSP Magazine, Vol. 4, Issue 2
Rescaling of variables in back propagation learning
journal, January 1991
- Rigler, A. K.; Irvine, J. M.; Vogl, T. P.
- Neural Networks, Vol. 4, Issue 2
ANFIS: adaptive-network-based fuzzy inference system
journal, January 1993
- Jang, J. -S. R.
- IEEE Transactions on Systems, Man, and Cybernetics, Vol. 23, Issue 3
Back-propagation algorithm which varies the number of hidden units
journal, January 1991
- Hirose, Yoshio; Yamashita, Koichi; Hijiya, Shimpei
- Neural Networks, Vol. 4, Issue 1
A prototype of neuro-fuzzy cooperation system
conference, January 1992
- Kawamura, A.; Watanabe, N.; Okada, H.
- [1992 Proceedings] IEEE International Conference on Fuzzy Systems
Fuzzy logic in control systems: fuzzy logic controller. II
journal, January 1990
- Lee, C. C.
- IEEE Transactions on Systems, Man, and Cybernetics, Vol. 20, Issue 2