用于光电共调制神经形态计算的氧化锌-氧化铟锡/氧化钨异质结忆阻器

扫码查看

原文链接

NETL
NSTL
万方数据

中文摘要：传统晶体管在大规模神经形态系统中面临计算能力不足和功耗过高的严峻挑战.为了解决这些关键问题,我们提出了一种基于氧化锌-氧化铟锡/氧化钨异质结的光电忆阻器,作为解决上述问题的有效方案.通过应用不同类型的电和光信号,该器件成功模拟了多种突触功能,包括短期/长期突触可塑性以及短期/长期记忆.引入氧化锌-氧化铟锡功能层增强了基于氧化钨忆阻器的光响应,并展示了光调制下的"学习-遗忘-再学习"行为.此外,在光电协同忆阻器阵列的基础上,我们构建了用于车型识别的卷积神经网络,解决了零权重和负权重实现复杂的问题.在功耗方面,使用该器件构建的神经网络仿真功率仅为10-3W,与标准中央处理器相比至少降低了4个数量级.因此,这项工作所提出的光电忆阻器为神经形态计算提供了新的思路,有望推动高能效类脑计算的发展.

外文标题：ZnO-ITO/WO3-x heterojunction structured memristor for optoelectronic co-modulation neuromorphic computation

外文摘要：Traditional transistors confront severe chal-lenges of insufficient computing capability and excessive power consumption in large-scale neuromorphic systems.To address these critical bottlenecks,we propose an optoelec-tronic memristor based on zinc oxide-indium tin oxide/tungsten oxide(ZnO-ITO/WO3-x)heterojunctions as a pro-mising solution.Through applying different types of electrical and optical signals,the device successfully emulates diverse synaptic functions including short-term/long-term synaptic plasticity,alongside short-term and long-term memory.In-troducing the ZnO-ITO functional layer enhances the photo-response of the WO3-x-based memristor and demonstrates"learning-forgetting-relearning"behavior under optical modulation.Furthermore,based on the photoelectric co-operative memristor array,a convolutional neural network for vehicle type recognition is constructed,which solves the pro-blem of zero weight and negative weight complexity.In regard to energy efficiency,the neural network built with this device operates at a power level of only 10-3 W,representing a re-duction of more than 4 orders of magnitude compared with a standard central processor.Hence,the photoelectric memris-tor proposed in this work provides a new idea for neuro-morphic computing and is expected to promote the development of energy-efficient brain-like computing.

外文关键词：

ZnO-ITO/WO3-x heterojunctionoptoelectronic memristorsynaptic plasticityneuromorphic computingcon-volutional neural network

作者：

潘建勇、吴彤、杨文豪、李阳、张佳旗、阚皞

展开 >

作者单位：

Shandong Provincial Key Laboratory of Network Based Intelligent Computing,School of Information Science and Engineering,University of Jinan,Jinan 250022,China

School of Integrated Circuits,Shandong University,Jinan 250101,China

Key Laboratory of Automobile Materials Ministry of Education,College of Materials Science and Engineering,Jilin University,Changchun 130012,China

关键词：

ZnO-ITO/WO3-x heterojunction optoelectronic memristor synaptic plasticity neuromorphic computing con-volutional neural network

基金：

National Natural Science Foundation of ChinaNational Natural Science Foundation of ChinaNational Natural Science Foundation of ChinaNational Natural Science Foundation of ChinaJinan City-University Integrated Development Strategy ProjectTaishan Scholars Project Special FundsNational Key Research and Development Program of Chinafunded by MOSTNational Natural Science Foundation of ChinaNatural Science Foundation of Jilin Province

项目编号：

62174068623115401556217406861804063JNSX2023017tsqn2023120352019YFA07059006180406320220201070GX

出版年：

2024

DOI：

10.1007/s40843-024-2946-y

中国科学:材料科学(英文)

CSTPCD

ISSN：

年,卷(期)：2024.67(9)