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

Adaptive model conversion method and system for spiking neural network

A technology of spiking neural network and self-adaptive model, applied in the field of neural network model, which can solve the problems of restricting the development and application of spiking neural network and inconvenience for developers

Pending Publication Date: 2021-01-29
NAT UNIV OF DEFENSE TECH
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, there are large differences in the deep artificial neural network models under different intelligent frameworks, and it is impossible to use the same method to convert multiple models to spiking neural networks at the same time, which greatly restricts the development and application of spiking neural networks. Developers have brought great inconvenience

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
  • Adaptive model conversion method and system for spiking neural network
  • Adaptive model conversion method and system for spiking neural network
  • Adaptive model conversion method and system for spiking neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0050] Such as figure 1 As shown, the self-adaptive model conversion method for the spiking neural network in this embodiment includes:

[0051] 1) obtain the depth artificial neural network model to be converted, the depth artificial neural network model to be converted is the depth neural network model under any framework in Keras, Pytorch, TensorFlow;

[0052] 2) If the deep artificial neural network model to be converted is a deep neural network model under the Pytorch or TensorFlow framework, it is converted to a deep artificial neural network model under the Keras framework;

[0053] 3) Convert the deep artificial neural network model under the Keras framework into a spiking neural network.

[0054] Convolutional neural network is a typical deep artificial neural network model. Here, it will be used as an example to describe the specific implementation of the technology. figure 2 It shows the process of adaptively converting a convolutional neural network model under...

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 adaptive model conversion method and system for a spiking neural network. The method comprises the steps: obtaining a to-be-converted deep artificial neural network model, and enabling the to-be-converted deep artificial neural network model to be a deep neural network model under any one of Keras, Pytorch and TensorFlow; if the to-be-converted deep artificial neural network model is a deep neural network model under a Pylot or TensorFlow framework, converting the to-be-converted deep artificial neural network model into a deep artificial neural network model under aKeras framework; converting the deep artificial neural network model under the Keras framework into an impulsive neural network. According to the method disclosed in the invention, the deep artificial neural network model under various intelligent frameworks can be realized, and the adaptive conversion to the novel pulse neural network is completed on the premise of ensuring a small amount of model precision loss.

Description

technical field [0001] The invention relates to a neural network model, in particular to an adaptive model conversion method and system for a pulse neural network. Background technique [0002] As a new-generation neural network model, the spiking neural network can capture the key dynamics of the input and output characteristics of biological neurons. Compared with traditional artificial neural networks, it is easier to build larger-scale networks while operating with low power consumption. Therefore, the spiking neural network model is a more promising brain-like computing model. [0003] Since the dynamics principle of the spiking neural network is more complex than that of the deep neural network, and the output state of the neuron is a step pulse signal, which itself does not have mathematical derivability, it is difficult to establish a gradient descent learning theory similar to the deep neural network. Therefore, for It is extremely challenging to directly learn and...

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): G06N3/04G06N3/063
CPCG06N3/063G06N3/045
Inventor 张毅刘晓东彭龙马俊黄辰林余杰谭郁松吴庆波廖湘科汪昌棋李文杰
Owner NAT UNIV OF DEFENSE TECH
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