Unlock instant, AI-driven research and patent intelligence for your innovation.

Fuzzy neural network learning algorithm based on loop learning

A technology of fuzzy neural network and learning algorithm, applied in the field of fuzzy neural network learning algorithm, to achieve the effect of improving classification accuracy, improving generalization ability and optimizing network structure

Inactive Publication Date: 2018-03-06
SHANGHAI DIANJI UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

On the other hand, if the value of the maximum scale limiting parameter θ is too small, many unnecessary hyperboxes will be built or produced

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Fuzzy neural network learning algorithm based on loop learning
  • Fuzzy neural network learning algorithm based on loop learning
  • Fuzzy neural network learning algorithm based on loop learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0048] Below in conjunction with specific embodiment, further illustrate the present invention. It should be understood that these examples are only used to illustrate the present invention and are not intended to limit the scope of the present invention. In addition, it should be understood that after reading the teachings of the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms also fall within the scope defined by the appended claims of the present application.

[0049] In the original FMM learning algorithm, all training samples are only passed through the learning algorithm once. When a training sample is applied to the FMM model, the model learns the sample sequentially: hyperbox expansion (or new creation), overlapping Steps like detection, superbox shrinkage (if needed), etc. All training modes train the network in turn, and when each training mode has been used once, the training...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to a fuzzy neural network learning algorithm based on loop learning. The algorithm includes the steps of calculating each hyperbox confidence value after expansion of each layerof hyperboxes through a defined "hyperbox confidence" value; and after each learning process of all training samples, analyzing the classification performance of all the existing hyperboxes, deletingthe "low confidence" hyperboxes, keeping the "high confidence" hyperboxes, then decreasing the maximum scale limit parameter value [theta], and performing training again until the performance of a classifier meets the requirements, or the minimum allowed value is reached. The fuzzy neural network learning algorithm in the invention can not only avoid the difficulty in defining the maximum scale limit parameter value [theta], but also limit the blind increase of the number of the hyperboxes in the expansion process of the hyperboxes.

Description

technical field [0001] The invention relates to a fuzzy neural network learning algorithm based on cycle learning. Background technique [0002] The fuzzy neural network is a model that attracts attention. The fuzzy minimum-maximum neural network (FMM) proposed by Simpso is a fuzzy neural network that uses a hyperbox membership function and can be applied to pattern classification, function approximation, and cluster analysis. problem, has been extensively studied recently and has applications in areas such as character recognition and detection. [0003] The learning algorithm of FMM consists of four steps: initialization-expansion-overlap test-compression. Each time a sample is input, the above four steps are repeated until the boundary is stable. In these four-step learning algorithms, the expansion criterion of the hyperbox (the second step of the learning algorithm) is determined by the scale limit parameters of the hyperbox, that is to say, the scale limit parameters...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06N3/04G06N3/08
CPCG06N3/08G06N3/043
Inventor 罗宜元
Owner SHANGHAI DIANJI UNIV