一种基于随机存储器的自适应阈值LIF神经元电路，具有较高的识别精度

2018 International Symposium on VLSI Technology, Systems and Application (VLSI-TSA) Pub Date : 2018-04-01 DOI:10.1109/VLSI-TSA.2018.8403854

Xinxin Wang, Peng Huang, Zhen Dong, Zheng Zhou, Yuning Jiang, Runze Han, Lifeng Liu, Xiaoyan Liu, Jinfeng Kang

{"title":"一种基于随机存储器的自适应阈值LIF神经元电路，具有较高的识别精度","authors":"Xinxin Wang, Peng Huang, Zhen Dong, Zheng Zhou, Yuning Jiang, Runze Han, Lifeng Liu, Xiaoyan Liu, Jinfeng Kang","doi":"10.1109/VLSI-TSA.2018.8403854","DOIUrl":null,"url":null,"abstract":"A novel leaky integrate-and-fire (LIF) neuron circuit based on the gradual switching in resistive random access memory (RRAM) device is put forward, in which threshold modulation can be achieved. Its threshold modulation and spike generating functions are verified through HSPICE simulation. In unsupervised pattern recognition for handwritten digits in MNIST dataset, its advantage in improving the accuracy (from about 70% to more than 95%) is demonstrated. Benchmarking results indicate that this novel neuron is much faster and can save about 66% area compared to one previously proposed.","PeriodicalId":209993,"journal":{"name":"2018 International Symposium on VLSI Technology, Systems and Application (VLSI-TSA)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"6","resultStr":"{\"title\":\"A novel RRAM-based adaptive-threshold LIF neuron circuit for high recognition accuracy\",\"authors\":\"Xinxin Wang, Peng Huang, Zhen Dong, Zheng Zhou, Yuning Jiang, Runze Han, Lifeng Liu, Xiaoyan Liu, Jinfeng Kang\",\"doi\":\"10.1109/VLSI-TSA.2018.8403854\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"A novel leaky integrate-and-fire (LIF) neuron circuit based on the gradual switching in resistive random access memory (RRAM) device is put forward, in which threshold modulation can be achieved. Its threshold modulation and spike generating functions are verified through HSPICE simulation. In unsupervised pattern recognition for handwritten digits in MNIST dataset, its advantage in improving the accuracy (from about 70% to more than 95%) is demonstrated. Benchmarking results indicate that this novel neuron is much faster and can save about 66% area compared to one previously proposed.\",\"PeriodicalId\":209993,\"journal\":{\"name\":\"2018 International Symposium on VLSI Technology, Systems and Application (VLSI-TSA)\",\"volume\":\"14 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2018-04-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"6\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2018 International Symposium on VLSI Technology, Systems and Application (VLSI-TSA)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/VLSI-TSA.2018.8403854\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 International Symposium on VLSI Technology, Systems and Application (VLSI-TSA)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/VLSI-TSA.2018.8403854","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 6

摘要

提出了一种基于电阻式随机存取存储器(RRAM)器件渐进开关的泄漏集成点火(LIF)神经元电路，该电路可实现阈值调制。通过HSPICE仿真验证了其阈值调制和尖峰产生功能。在对MNIST数据集中手写数字的无监督模式识别中，证明了其在提高准确率方面的优势(从70%左右提高到95%以上)。基准测试结果表明，与之前提出的神经元相比，这种新型神经元的速度要快得多，可以节省约66%的面积。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

A novel RRAM-based adaptive-threshold LIF neuron circuit for high recognition accuracy

A novel leaky integrate-and-fire (LIF) neuron circuit based on the gradual switching in resistive random access memory (RRAM) device is put forward, in which threshold modulation can be achieved. Its threshold modulation and spike generating functions are verified through HSPICE simulation. In unsupervised pattern recognition for handwritten digits in MNIST dataset, its advantage in improving the accuracy (from about 70% to more than 95%) is demonstrated. Benchmarking results indicate that this novel neuron is much faster and can save about 66% area compared to one previously proposed.

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

2018 International Symposium on VLSI Technology, Systems and Application (VLSI-TSA)

自引率

0.00%

发文量