基于 Memristor 电子元件基础的神经形态电子模块用于图像识别

IF 0.8 Q3 Engineering

Nanotechnologies in Russia Pub Date : 2024-03-21 DOI:10.1134/S2635167623600724

E. A. Ryndin, I. A. Mavrin, N. V. Andreeva, V. V. Luchinin

{"title":"基于 Memristor 电子元件基础的神经形态电子模块用于图像识别","authors":"E. A. Ryndin, I. A. Mavrin, N. V. Andreeva, V. V. Luchinin","doi":"10.1134/S2635167623600724","DOIUrl":null,"url":null,"abstract":"<div><p>In this work a method for increasing the efficiency of hardware implementation and minimizing the number of electronic synaptic elements of asynchronous spiking neural networks in solving image identification problems is developed. The effectiveness of the proposed method is demonstrated in the process of optimizing the parameters of neurons and training the neural network on the developed software model and confirmed by the results of SPICE modeling and measurement of the signals of the neural network implemented on series electronic components.</p></div>","PeriodicalId":716,"journal":{"name":"Nanotechnologies in Russia","volume":"18 1 supplement","pages":"S194 - S202"},"PeriodicalIF":0.8000,"publicationDate":"2024-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Neuromorphic Electronic Module Based on the Use of the Memristor Electronic-Component Base for Image Recognition\",\"authors\":\"E. A. Ryndin, I. A. Mavrin, N. V. Andreeva, V. V. Luchinin\",\"doi\":\"10.1134/S2635167623600724\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>In this work a method for increasing the efficiency of hardware implementation and minimizing the number of electronic synaptic elements of asynchronous spiking neural networks in solving image identification problems is developed. The effectiveness of the proposed method is demonstrated in the process of optimizing the parameters of neurons and training the neural network on the developed software model and confirmed by the results of SPICE modeling and measurement of the signals of the neural network implemented on series electronic components.</p></div>\",\"PeriodicalId\":716,\"journal\":{\"name\":\"Nanotechnologies in Russia\",\"volume\":\"18 1 supplement\",\"pages\":\"S194 - S202\"},\"PeriodicalIF\":0.8000,\"publicationDate\":\"2024-03-21\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Nanotechnologies in Russia\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://link.springer.com/article/10.1134/S2635167623600724\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"Engineering\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Nanotechnologies in Russia","FirstCategoryId":"1085","ListUrlMain":"https://link.springer.com/article/10.1134/S2635167623600724","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"Engineering","Score":null,"Total":0}

引用次数: 0

摘要

在这项工作中，开发了一种在解决图像识别问题时提高硬件实施效率和尽量减少异步尖峰神经网络电子突触元件数量的方法。在优化神经元参数和在开发的软件模型上训练神经网络的过程中，证明了所提方法的有效性，并通过在串联电子元件上实施的神经网络信号的 SPICE 建模和测量结果得到了证实。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

Neuromorphic Electronic Module Based on the Use of the Memristor Electronic-Component Base for Image Recognition

查看原文本刊更多论文

Neuromorphic Electronic Module Based on the Use of the Memristor Electronic-Component Base for Image Recognition

In this work a method for increasing the efficiency of hardware implementation and minimizing the number of electronic synaptic elements of asynchronous spiking neural networks in solving image identification problems is developed. The effectiveness of the proposed method is demonstrated in the process of optimizing the parameters of neurons and training the neural network on the developed software model and confirmed by the results of SPICE modeling and measurement of the signals of the neural network implemented on series electronic components.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Nanotechnologies in Russia NANOSCIENCE & NANOTECHNOLOGY-

CiteScore

1.20

自引率

0.00%

发文量

期刊介绍： Nanobiotechnology Reports publishes interdisciplinary research articles on fundamental aspects of the structure and properties of nanoscale objects and nanomaterials, polymeric and bioorganic molecules, and supramolecular and biohybrid complexes, as well as articles that discuss technologies for their preparation and processing, and practical implementation of products, devices, and nature-like systems based on them. The journal publishes original articles and reviews that meet the highest scientific quality standards in the following areas of science and technology studies: self-organizing structures and nanoassemblies; nanostructures, including nanotubes; functional and structural nanomaterials; polymeric, bioorganic, and hybrid nanomaterials; devices and products based on nanomaterials and nanotechnology; nanobiology and genetics, and omics technologies; nanobiomedicine and nanopharmaceutics; nanoelectronics and neuromorphic computing systems; neurocognitive systems and technologies; nanophotonics; natural science methods in a study of cultural heritage items; metrology, standardization, and monitoring in nanotechnology.