Construction and optimization method of memristor array neural network based on circuit simulation
A circuit simulation and neural network technology, which is applied in the construction and optimization of memristor array neural network, can solve problems such as data disturbance and calculation result error, and achieve high accuracy
Active Publication Date: 2021-10-29
SOUTH UNIVERSITY OF SCIENCE AND TECHNOLOGY OF CHINA
View PDF5 Cites 6 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] The technical problem to be solved by the present invention is to provide a method for constructing and optimizing the memristor array neural network based on circuit simulation, aiming at solving the problems in the working process of the memristor in the prior art. In , some non-ideal physical characteristics, such as read and write noise, random effects, and release effects, etc., will disturb the data of its calculations, resulting in errors from the ideal calculation results.
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0056]The invention discloses a method for constructing and optimizing a memristor array neural network based on circuit simulation. In order to make the purpose, technical solution and effect of the invention clearer and clearer, the invention will be further described in detail below with reference to the accompanying drawings and examples . It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.
[0057] Those skilled in the art will understand that unless otherwise stated, the singular forms "a", "an", "said" and "the" used herein may also include plural forms. It should be further understood that the word "comprising" used in the description of the present invention refers to the presence of said features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, Integers, steps, operations, elements,...
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
Login to View More Abstract
The invention discloses a construction and optimization method of a memristor array neural network based on circuit simulation. The method comprises the steps of obtaining a trained neural network model; obtaining a memristor array neural network model according to the trained neural network model and a preset memristor; and based on circuit simulation, optimizing the memristor array neural network model to obtain an optimization result of the memristor array neural network model. According to the embodiment of the invention, the neural network model is combined with the memristor array to obtain the memristor array neural network model, and the memristor array neural network model is optimized, so that the accuracy of data processing can be improved.
Description
technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to a method for constructing and optimizing a memristor array neural network based on circuit simulation. Background technique [0002] Traditional computers based on the Von Neumann hardware architecture face the serious restriction of the "Memory Wall" when performing interactive calculations of large amounts of data, that is, the access speed of the memory lags behind the computing speed of the processor. Especially during neural network training, the huge data exchange between the processor and the memory will generate huge energy consumption, and the calculation speed is slow, which cannot effectively process real-time data. To fundamentally solve the problem, new computing hardware technology is needed. Memristors (memristors) are widely considered to be the next generation of computing hardware that can solve the von Neumann bottleneck. [0003] Existing mem...
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
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06N3/063G06N3/08
CPCG06N3/063G06N3/084Y02T10/40
Inventor 陈全范达熠谢锐宋明洋
Owner SOUTH UNIVERSITY OF SCIENCE AND TECHNOLOGY OF CHINA