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

A Probabilistic Determination Method for Nondeterministic Problems

A non-deterministic and deterministic technology, applied in neural learning methods, biological neural network models, neural architectures, etc., and can solve problems such as limited application of neuron circuits

Active Publication Date: 2022-08-05
HUAZHONG UNIV OF SCI & TECH
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The present invention provides a probabilistic neuron circuit, a probabilistic neural network topology and its application, which are used to solve the technical problem that the application of neuron circuits is limited due to the use of deterministic neurons in existing neuron circuits

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 Probabilistic Determination Method for Nondeterministic Problems
  • A Probabilistic Determination Method for Nondeterministic Problems
  • A Probabilistic Determination Method for Nondeterministic Problems

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] A probabilistic neuron circuit 100, such as figure 1 As shown, including: integrating capacitor, non-fixed threshold volatile device and load resistance; one end of the integrating capacitor is grounded, the other end is connected to a synaptic resistor to connect an external signal input source and one end of the non-fixed threshold volatile device, non-fixed The other end of the threshold volatile device is connected to one end of the load resistor, and the other end of the load resistor is grounded.

[0045] Due to the characteristics of the non-fixed threshold volatile device, its excitation (turn-on) voltage threshold is not fixed and is a randomly changing value. The probability corresponding to each excitation voltage is generally different, and the variation law roughly satisfies the Gaussian distribution. figure 2 As shown, the voltage required for a volatile memory device to change from a high-resistance state to a low-resistance state becomes the turn-on vol...

Embodiment 2

[0049] A probabilistic neural network topology 200, such as image 3 As shown, it includes: a plurality of input neuron circuits, a plurality of output neuron circuits, a lateral inhibitory neuron circuit, and a signal processor; wherein, the output neuron circuits are the probabilistic neuron circuits described above;

[0050] Each input neuron circuit is used to send a discharge signal to each probability neuron circuit; each probability neuron circuit is used for random excitation based on its non-fixed excitation threshold and the electrical signal emitted by each input neuron circuit; lateral inhibition The neuron circuit is used to inhibit the subsequent excitation of other probability neuron circuits when receiving the signals excited by the first n probability neuron circuits, where n≥1; the signal processor is used to collect whether the excitation of each probability neuron circuit is not signal and perform signal processing.

[0051] A plurality of input neuron cir...

Embodiment 3

[0060] An application of any of the probabilistic neural network topology structures described in the second embodiment above is applied to probability determination of non-deterministic problems.

[0061] It should be noted that the non-deterministic problem is called Uncertainty Quantification.

[0062] Based on the foregoing, the excitation threshold of the non-fixed threshold volatile device is not fixed and the excitation is random, so the excitation has uncertainty, not the higher the membrane voltage (turn-on voltage) of the non-fixed threshold volatile device. Exciting first, this is a probabilistic event. Therefore, using non-fixed threshold volatile devices for probability determination of non-deterministic problems, using hardware on natural properties, the probability generated is with real probability properties, and at this stage, computer In , the method of generating probability is to generate a random number, and then realize it by mathematical algorithm, and ...

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 a probabilistic neuron circuit, a topological structure of a probabilistic neural network and an application thereof. The probabilistic neuron circuit comprises an integral capacitor, a non-fixed threshold volatile device and a load resistor; One end of the fixed-threshold volatile device, and the other end of the volatile device is connected to one end of the load resistor. The network topology includes multiple input neuron circuits, multiple probabilistic neuron circuits, and lateral inhibitory neuron circuits; each probabilistic neuron circuit is used to perform a neural network based on its non-fixed excitation threshold and the electrical signals emitted by each input neuron circuit. Random excitation; the inhibitory neuron circuit is used to inhibit the excitation of other subsequent probabilistic neuron circuits when receiving the signals excited by the first n probability neuron circuits. The invention introduces a non-fixed threshold volatile device into the neuron circuit, which greatly expands the application of the neuron circuit, especially can be used to solve non-deterministic problems, and has reliable solution results.

Description

technical field [0001] The invention belongs to the technical field of microelectronic devices, and more particularly, relates to a probabilistic neuron circuit, a probabilistic neural network topology and applications thereof. Background technique [0002] By simulating the learning principle of the human brain, brain-like computing has the characteristics of high speed, low power consumption and self-learning, and is a strong competitor to replace the current von Neumann computing architecture. The core mechanism of brain-like computing is to simulate the human brain to emit impulses through the excitation of neurons, complete the transmission of information, and then adjust the weight of the synaptic connection between the anterior and posterior neurons to complete the learning function. The network with pulses as the information transmission carrier is called spiking neural network. At the hardware level, microelectronic devices are used to simulate the functions of neur...

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): G06N3/04G06N3/063G06N3/08
CPCG06N3/049G06N3/063G06N3/08
Inventor 童浩胡庆王宽何毓辉缪向水
Owner HUAZHONG UNIV OF SCI & 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