期刊文献+

A physical memristor model for Pavlovian associative memory

在线阅读 下载PDF
导出
摘要 Brain-inspired intelligence is considered to be a computational model with the most promising potential to overcome the shortcomings of the von Neumann architecture,making it a current research hotspot.Due to advantages such as nonvolatility,high density,low power consumption,and high response ratio,memristors are regarded as devices with promising applications in brain-inspired intelligence.This paper proposes a physical Ag/HfO_(x)/FeO_(x)/Pt memristor model.The Ag/HfO_(x)/FeO_(x)/Pt memristor is first fabricated using magnetron sputtering,and its internal principles and characteristics are then thoroughly analyzed.Furthermore,we construct a corresponding physical memristor model which achieves a simulation accuracy of up to 99.72%for the physical memristor.We design a fully functional Pavlovian associative memory circuit,realizing functions including generalization,primary differentiation,secondary differentiation,and forgetting.Finally,the circuit is validated through PSPICE simulation and analysis.
作者 Jiale Lu Haofeng Ran Dirui Xie Guangdong Zhou Xiaofang Hu 卢家乐;冉皓丰;谢頔睿;周广东;胡小方(College of Artificial Intelligence,Southwest University,Chongqing 400712,China)
出处 《Chinese Physics B》 2025年第1期507-517,共11页 中国物理B(英文版)
基金 Project supported by the National Natural Science Foundation of China(Grant Nos.62476230 and 61976246) the Natural Science Foundation of Chongqing(Grant No.CSTB2023NSCQ-MSX0018) Fundamental Research Funds for the Central Universities(Grant No.SWUKR22046)。
作者简介 Corresponding author: Jiale Lu,E-mail: krislu 111@163.com;Corresponding author: Guangdong Zhou,E-mail: zhougd@swu.edu.cn;Corresponding author: Xiaofang Hu,E-mail: huxf@swu.edu.cn。
  • 相关文献

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部