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

A method and device for classification of iris flowers based on evolvable spiking neural network

A technology of pulse neural network and classification method, which is applied in the field of iris flower classification, can solve the problem of no effective algorithm, etc., and achieve the effect of avoiding over-training, speeding up the evolution of the network, and avoiding unnecessary updates

Active Publication Date: 2020-10-16
SHANDONG FIRST MEDICAL UNIV & SHANDONG ACADEMY OF MEDICAL SCI
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The current research on spiking neural networks is in its infancy, and there is still a lack of learning algorithms that conform to biological mechanisms. Neural networks should be learning and evolutionary, but there is no effective algorithm in this regard

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 method and device for classification of iris flowers based on evolvable spiking neural network
  • A method and device for classification of iris flowers based on evolvable spiking neural network
  • A method and device for classification of iris flowers based on evolvable spiking neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0041] This embodiment discloses a method for constructing an evolvable spiking neural network, including:

[0042] Step 1: Initialize the evolvable spiking neural network;

[0043] Specifically, the neural network includes an input layer, a supervisory neuron and an output layer; the maximum number of output neurons of the network is set, and the number of output types is determined according to the input training samples; and other parameters in the evolvable spiking neural network are initialized at the same time.

[0044] Such as figure 1 As shown, the structure is divided into input layer, supervisory neuron and output layer, and the input of spiking neural network is x={x 1 ,...,x i ,...,x m}, the output is Y={Y 1 ,...,Y j ,...,Y k}, the supervisory neuron is located between the input layer and the output layer, and is used to update the network weights according to the existing characteristics of the network. The supervisory neuron provides a correction parameter...

Embodiment 2

[0084] The purpose of this embodiment is to provide a computing device.

[0085] An evolvable spiking neural network construction device, comprising a memory, a processor, and a computer program stored on the memory and operable on the processor, and the processor implements the following steps when executing the program, including:

[0086] Step 1: Initialize the evolvable spiking neural network;

[0087] Step 2: Use the training samples to train the evolvable spiking neural network, and calculate the corresponding moment when the sample generates a spike in the post-synaptic potential time region;

[0088] Step 3: Evolve the neural network based on the moment selection network evolution strategy, the network evolution strategy includes adding output neuron strategy, canceling input pulse sequence training strategy and weight parameter updating strategy.

Embodiment 3

[0090] The purpose of this embodiment is to provide a computer-readable storage medium.

[0091] A computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the following steps are performed:

[0092] Step 1: Initialize the evolvable spiking neural network;

[0093] Step 2: Use the training samples to train the evolvable spiking neural network, and calculate the corresponding moment when the sample generates a spike in the post-synaptic potential time region;

[0094] Step 3: Evolve the neural network based on the moment selection network evolution strategy, the network evolution strategy includes adding output neuron strategy, canceling input pulse sequence training strategy and weight parameter updating strategy.

[0095] The steps involved in the devices of the above embodiments 2 and 3 correspond to those of the method embodiment 1, and for specific implementation methods, please refer to the relevant descrip...

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 discloses an evolvable pulse neural network construction method and device. The method comprises the steps of initializing an evolvable pulse neural network; training the evolvable pulseneural network by using training samples, calculating the corresponding moment of generating a pulse in a potential time region after the samples are subjected to the sudden contact, and selecting network evolution strategies based on the pulse generation time to evolve the neural network, wherein the network evolution strategies comprise an output neuron adding strategy, an input pulse sequencecanceling training strategy and a weight parameter updating strategy. By means of selection of three evolutionary strategies, unnecessary updating of network weights can be avoided, and network evolution speed is increased.

Description

technical field [0001] The invention belongs to the field of pulse neural network construction, and in particular relates to an iris flower classification method and device based on an evolvable pulse neural network. Background technique [0002] Artificial neural network is a technology widely used in the field of natural science. So far, artificial neural network has experienced three generations of development. The first-generation artificial neural network is formed based on the McCulloch-Pitts neuron model, and the output is a Boolean logic variable; the second-generation artificial neural network uses a continuous function as the activation function to suit the realization of the analog input / output of the system. However, some studies in recent years have shown that the method of frequency encoding in biological systems is often not applicable. The researchers found that neurons in the cerebral cortex can transmit information at an incredible speed, and the assumptio...

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
IPC IPC(8): G06N3/08
CPCG06N3/086
Inventor 陆强田娟张兆臣李文锋
Owner SHANDONG FIRST MEDICAL UNIV & SHANDONG ACADEMY OF MEDICAL SCI
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