# 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{minus}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 shown in Equation 1 where N{sub k{minus}1} is the number of nodes in layer k{minus}1, i indexes the nodes of layer k{minus}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{sub 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{minus}1} in layer n{minus}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 Equation 2, shown in the patent. 8 figs.

- Inventors:

- Issue Date:

- Research Org.:
- Univ. of California (United States)

- OSTI Identifier:
- 527782

- Patent Number(s):
- 5,664,066

- Application Number:
- PAN: 8-304,660

- Assignee:
- Dept. of Energy, Washington, DC (United States)

- DOE Contract Number:
- W-7405-ENG-48

- Resource Type:
- Patent

- Resource Relation:
- Other Information: PBD: 2 Sep 1997

- Country of Publication:
- United States

- Language:
- English

- Subject:
- 99 MATHEMATICS, COMPUTERS, INFORMATION SCIENCE, MANAGEMENT, LAW, MISCELLANEOUS; NEURAL NETWORKS; IDENTIFICATION SYSTEMS; FUZZY LOGIC; TRAINING; WEIGHTING FUNCTIONS

### Citation Formats

```
Sun, C T, Shiang, P S, Jang, J S, and Fu, C Y.
```*Intelligent system for automatic feature detection and selection or identification*. United States: N. p., 1997.
Web.

```
Sun, C T, Shiang, P S, Jang, J S, & Fu, C Y.
```*Intelligent system for automatic feature detection and selection or identification*. United States.

```
Sun, C T, Shiang, P S, Jang, J S, and Fu, C Y. Tue .
"Intelligent system for automatic feature detection and selection or identification". United States.
```

```
@article{osti_527782,
```

title = {Intelligent system for automatic feature detection and selection or identification},

author = {Sun, C T and Shiang, P S and Jang, J S and Fu, C Y},

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{minus}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 shown in Equation 1 where N{sub k{minus}1} is the number of nodes in layer k{minus}1, i indexes the nodes of layer k{minus}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{sub 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{minus}1} in layer n{minus}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 Equation 2, shown in the patent. 8 figs.},

doi = {},

journal = {},

number = ,

volume = ,

place = {United States},

year = {1997},

month = {9}

}