Tunneling magnetoresistance materials and devices for neuromorphic computing
作者机构：Fert Beijing InstituteMIIT Key Laboratory of SpintronicsSchool of Integrated Circuit Science and EngineeringBeihang UniversityBeijing 100191People’s Republic of China Hefei Innovation Research InstituteAnhui High Reliability Chips Engineering LaboratoryBeihang UniversityHefei 230013People’s Republic of China Truth Equipment CorporationTRECHefei 230013People’s Republic of China Beihang-Geortek Joint Microelectronics InstituteQingdao Research InstituteBeihang UniversityQingdao 266061People’s Republic of China
出 版 物：《Materials Futures》 (材料展望（英文）)
年 卷 期：2023年第2卷第3期
页 面：90-113页
核心收录：
学科分类：080903[工学-微电子学与固体电子学] 0809[工学-电子科学与技术（可授工学、理学学位）] 08[工学]
基 金：the National Key Research and Development Program of China(Grant Nos.2022YFB4400200 and 2022YFA1402604) the National Natural Science Foundation of China(Grant Nos.12104031 and 52121001) Science and Technology Major Project of Anhui Province(Grant No.202003a05020050) the International Collaboration Project B16001,the Beihang Hefei Innovation Research Institute Project BHKX-19-02,the China Postdoctoral Science Foundation No.2022M720345 Outstanding Research Project of Shenyuan Honors College BUAA 230121102 for their financial support of this work
主 题：spintronics TMR MTJ neuromorphic computing SNN
摘 要：Artificial intelligence has become indispensable in modern life,but its energy consumption has become a significant concern due to its huge storage and computational *** intelligence algorithms are mainly based on deep learning algorithms,relying on the backpropagation of convolutional neural networks or binary neural *** these algorithms aim to simulate the learning process of the human brain,their low bio-fidelity and the separation of storage and computing units lead to significant energy *** human brain is a remarkable computing machine with extraordinary capabilities for recognizing and processing complex information while consuming very low *** magnetoresistance(TMR)-based devices,namely magnetic tunnel junctions(MTJs),have great advantages in simulating the behavior of biological synapses and *** is not only because MTJs can simulate biological behavior such as spike-timing dependence plasticity and leaky integrate-fire,but also because MTJs have intrinsic stochastic and oscillatory *** characteristics improve MTJs’bio-fidelity and reduce their power *** also possess advantages such as ultrafast dynamics and non-volatile properties,making them widely utilized in the field of neuromorphic computing in recent *** conducted a comprehensive review of the development history and underlying principles of TMR,including a detailed introduction to the material and magnetic properties of MTJs and their temperature *** also explored various writing methods of MTJs and their potential ***,we provided a thorough analysis of the characteristics and potential applications of different types of MTJs for neuromorphic ***-based devices have demonstrated promising potential for broad application in neuromorphic computing,particularly in the development of spiking neural *** ability to perform on-chip learning with ultra-low power consumpt
维普期刊数据库