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

Switched artificial synapse

A technology of synapses and synaptic weights, applied in digital memory information, instruments, biological neural network models, etc., can solve the problems of increasing power consumption, unable to solve static consumption, and difficult to implement networks with complex architectures, to promote low power consumption. effect of reducing power consumption, reducing the risk of action potentials

Pending Publication Date: 2021-03-30
UNIV DE LILLE +4
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This generation of action potentials has two major disadvantages: (i) first, these action potentials are of no benefit in terms of useful information to the neural network; (ii) second, they increase power consumption
However, this solution is expensive in terms of implantation, requiring an additional transistor for each post-neuron, and this solution may be difficult to implement for networks with complex architectures
Also, it doesn't solve the problems related to static consumption

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
  • Switched artificial synapse
  • Switched artificial synapse
  • Switched artificial synapse

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0080] figure 2 The principle of interconnection between the synapse S and the pre-neuron Pre-N before it and the post-neuron Post-N after it is shown. For example is supplied with the voltage V dd This neural circuit 1 forms the basis of neural networks.

[0081] Pre-neurons and post-neurons can be of Morris-Lecar or Axon-Hillock type. In both cases, the circuit consists of cascaded at the output to generate the output voltage VA respectively out_bar and VA out The two inverters 2 and 3, such as Figure 3A or Figure 4 shown.

[0082] Figure 3A shown includes a film capacitance known as C mem Example of a Morris-Lecar neuronal nucleus "ML-N-C" of a capacitor whose terminals define the membrane potential V mem . ML-N-C also includes: two transistors PMOS 4 and NMOS 5 connected in series, their drain current I Na and I K Corresponding to the biological sodium current and potassium current, respectively; between the gate of the NMOS transistor 5 and the ground, a cap...

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 an electronic neural circuit comprising at least one pre-neuron having an output voltage VAout and at least one post-neuron which are connected by at least one excitatory synapse having at least one switching input, in which the excitatory synapse is powered by the output VAout and receives a switching signal VAout_bar on its switching input, the state of the switching signal being complementary to that of the output VAout.

Description

technical field [0001] The present invention relates to a circuit with low energy consumption capable of reproducing certain electrical properties of biological synapses and which can be used in particular in biomimetic architectures. [0002] More precisely, the present invention relates to novel artificial synaptic switch topologies with low power consumption, which can be associated with energy-efficient artificial neurons. Background technique [0003] Many works have been done to model the biological behavior of neurons, such as two IEEE articles: "Neuromorphic electronic systems" by C. Mead et al. 1990 and "Abio-physically inspired silicon Neuron". [0004] Application FR 3 050 050 describes a synaptic circuit using conventional CMOS technology associated with an artificial neuron whose energy consumption is approximately fJ per action potential. [0005] J.Arthur and K.Boahen proposed a CMOS circuit that simulates the plasticity of synapses in the article "Learning ...

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): G06N3/063G06N3/04
CPCG06N3/049G06N3/065G06N3/088G06N3/044G11C11/41G11C11/54G06N3/06
Inventor 弗朗索瓦·丹内维尔阿兰·卡佩伊利亚斯·索里科普洛斯克里斯多夫·卢瓦耶
Owner UNIV DE LILLE
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