Event-driven spiking neuron simulation algorithm based on single exponential kernel

An event-driven, neuron technology, applied in the field of brain-like computing and neuron model, can solve problems such as the efficiency gap of deep learning, and achieve the effect of improving recognition accuracy and robustness, reducing delay and improving efficiency.

Pending Publication Date: 2020-04-10
TIANJIN UNIV
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Therefore, compared with the superior cognitive ability of the brain, there is still a huge gap in the efficiency of deep learning

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
  • Event-driven spiking neuron simulation algorithm based on single exponential kernel
  • Event-driven spiking neuron simulation algorithm based on single exponential kernel
  • Event-driven spiking neuron simulation algorithm based on single exponential kernel

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0039] The use of the present invention will be described in detail below in conjunction with the accompanying drawings.

[0040] Based on the single exponential kernel function, the present invention proposes a more efficient and biologically reasonable LIF spiking neuron model. The neuron model is shown below.

[0041]

[0042] τ represents the time constant of the neuronal membrane potential. I in and I out Respectively represent the input current of the presynaptic neuron and the reset current after the neuron fires a pulse. Whenever the neuron fires a pulse, the neuron will have a corresponding reset dynamic response. I in and I out is defined as follows.

[0043]

[0044]

[0045] δ(t) is a unit pulse function, its value is 1 only when t=0, and its value is 0 at other times. is the time to reach the jth pulse at the ith synapse, Represents the time of the jth output pulse of the current neuron. N and w i Indicates the number of presynaptic neurons an...

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 provides an event-driven spiking neuron simulation algorithm based on a single exponential kernel. Compared with a traditional spiking neuron model, according to the algorithm, a singleexponential kernel function is used, the efficiency of the spiking neuron model is greatly improved, the delay degree of the spiking neuron model to input information is reduced, the recognition accuracy and robustness of the spiking neuron model are improved, and the method is more suitable for development and application of software and hardware platforms of brain-imitated architecture. Meanwhile, a method based on event driving is adopted, calculation is driven by pulses, and the method can efficiently process pulse modes.

Description

technical field [0001] The invention belongs to the field of brain-like computing and neuron models, and in particular relates to a technique for improving the computing performance of a spiking neuron model, in particular to an event-driven spiking neuron simulation algorithm based on a single exponential kernel. Background technique [0002] The human brain has demonstrated remarkable capabilities in various cognitive tasks such as recognition, decision-making, learning, and memory while consuming very low resources. The outstanding performance of the human brain has inspired more and more scientific researchers to try to understand its operating principles, and hope to apply these principles to artificial intelligence systems, so that they have the ability to process information similar to the human brain. [0003] Driven by deep learning technology, artificial neural networks have achieved great success in solving problems in various fields, including image and speech re...

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): G06N3/04G06N3/063
CPCG06N3/049G06N3/063
Inventor 于强宋世明李盛兰王龙标党建武
Owner TIANJIN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products