Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A wta neural network based on memristor array and its application

A neural network and memristor technology, applied in the field of neural networks, can solve the problems of large size, high power consumption, and unfavorable large-scale use of neural networks, and achieve the effects of small size, low power consumption, and excellent performance.

Inactive Publication Date: 2017-08-11
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Generally speaking, MOSFETs or CMOS are commonly used to physically implement neurons, and the resulting neural network is large in size and consumes a lot of power, which is not conducive to large-scale use.

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
  • A wta neural network based on memristor array and its application
  • A wta neural network based on memristor array and its application
  • A wta neural network based on memristor array and its application

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0030] Use 12 memristors, 12 MOS tubes, 4 capacitors, 4 resistors, and 4 DC power supplies to connect according to the above-mentioned connection method, that is, to construct a memristor-based with four neurons Array of WTA neural networks, such as figure 2 Shown. Analyzing the network, the dynamic equation (4) can be obtained, where N=4. Through mathematical analysis of the dynamic equation, it can be verified that the neural network has WTA characteristics.

Embodiment 2

[0032] Use 56 memristors, 56 MOS tubes, 8 capacitors, 8 resistors, and 8 DC power supplies to connect according to the above-mentioned connection method, that is, to construct a memristor-based with eight neurons Array of WTA neural network. Analyzing the network, the dynamic equation formula (4) can be obtained, where N=8. Through mathematical analysis of the dynamic equation, it can be verified that the neural network has WTA characteristics.

[0033] Finally, the MATLAB simulations of Example 1 and Example 2 were performed respectively, and the simulation results were as follows: image 3 , Figure 4 Shown. It can be seen from the simulation results that only one output is positive and the other outputs are negative, showing the WTA characteristics.

Embodiment 3

[0035] The WTA neural network classifier based on the memristor array combines the BP neural network with the WTA neural network based on the memristor array, which is connected through a converter. After the output of the first part of the BP network is isolated and driven by the second part of the converter, it is used as the input of the third part of the WTA neural network. Combine the above-mentioned embodiment 1 to construct a classifier, its structure is as follows Figure 5 Shown.

[0036] In order to verify the effect of the classifier, we apply it to the classification and recognition of skin diseases. We use the data in the Dermatology Database to classify the sub-types of skin diseases such as erythemato-scuamous. We used 366 groups of data: 72 groups were lichen planus (lichen planus), 49 groups were pityriasis rosea (pityriasis rosea), 52 groups were chronic dermatitis (chronic dermatitis), 112 groups were psoriasis (psoriasis), 61 groups were Seborrheic dermatiti...

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 belongs to the technical field of neural networks, and relates to a WTA neural network based on a memristor array and an application thereof. The WTA neural network and its application, the classifier model, are realized through the design of the memristor array, and a classification recognition method that can be used for the characteristics of skin diseases is proposed. Among them, the WTA model is composed of memristors, MOSFETs, capacitors, resistors and power supplies. On this basis, a classifier based on memristor array WTA neural network is derived. The purpose is to use memristors to realize WTA neural network and its classifier, and to explore the application of memristors in neural networks and medical decision-making. Compared with the traditional WTA neural network, the classifier designed by using the memristor array of the present invention has excellent performance, and can be applied to the fields of medical disease classification and the like.

Description

Technical field [0001] The invention relates to a memristor and a WTA neural network, provides a WTA neural network based on a memristor array, and designs an application based on the network. It belongs to the field of neural network technology. Background technique [0002] Artificial neural network is a computational model that abstracts the processing of information by a human brain neuron network. The main types of neural networks studied include BP neural network, cellular neural network, recurrent neural network, WTA neural network, etc. Among them, WTA neural network continues to expand in theoretical research and engineering applications, and shows good intelligent characteristics in the fields of pattern recognition, image processing, and automatic control. [0003] Artificial neural network is composed of a large number of interconnected neurons. Generally speaking, MOSFET or CMOS are commonly used to physically realize neurons, and the neural network formed therefrom...

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 Patents(China)
IPC IPC(8): G06N3/063
Inventor 于永斌刘兴文胡青青门乐飞杨辰宇李成邓建华张容权蔡竟业
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com