Configurable NbOx Memristors as Artificial Synapses or Neurons Achieved by Regulating the Forming Compliance Current for the Spiking Neural Network
Abstract For the first time, a configurable NbOx memristor is achieved that can be configured as an artificial synapse or neuron after fabrication by controlling the forming compliance current (FCC). When the FCC ≤ 2 mA, the memristors exhibit the resistive‐switching (RS) property, enabling multiple...
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley-VCH
2023-06-01
|
| Series: | Advanced Electronic Materials |
| Subjects: | |
| Online Access: | https://doi.org/10.1002/aelm.202300018 |
| _version_ | 1797672126456528896 |
|---|---|
| author | Chuan Yu Han Sheng Li Fang Yi Lin Cui Weihua Liu Shi Quan Fan Xiao Dong Huang Xin Li Xiao Li Wang Guo He Zhang Wing Man Tang P. T. Lai Jia Liu Xianjie Wan Zhou Yu Li Geng |
| author_facet | Chuan Yu Han Sheng Li Fang Yi Lin Cui Weihua Liu Shi Quan Fan Xiao Dong Huang Xin Li Xiao Li Wang Guo He Zhang Wing Man Tang P. T. Lai Jia Liu Xianjie Wan Zhou Yu Li Geng |
| author_sort | Chuan Yu Han |
| collection | DOAJ |
| description | Abstract For the first time, a configurable NbOx memristor is achieved that can be configured as an artificial synapse or neuron after fabrication by controlling the forming compliance current (FCC). When the FCC ≤ 2 mA, the memristors exhibit the resistive‐switching (RS) property, enabling multiple types of synaptic plasticity, including short‐term potentiation, paired‐pulse facilitation, short‐term memory, and long‐term memory. When the FCC ≥ 3 mA, the memristors can be electroformed and exhibit the threshold switching (TS) property with excellent endurance (>1012), thus achieving various biological neuron characteristics, such as threshold‐triggering, strength‐modulation of spike frequency, and leaky integrate‐and‐fire. This enables the successful implementation of a spiking Pavlov's dog that employs the spikes as information carrier by connecting an RS NbOx memristor as artificial synapse and a TS memristor as artificial neuron in series. Furthermore, a fully NbOx memristors‐based single‐layer spiking neural network is simulated. It is first found that, due to the forgetting property of synapse, the recognition accuracy for the Modified National Institute of Standards and Technology handwritten digits is increased from 85.49% to 91.45%. This study provides a solid foundation for the development of neuromorphic machines based on the principles of the human brain. |
| first_indexed | 2024-03-11T21:25:37Z |
| format | Article |
| id | doaj.art-1e1f0d6a7df54916a6ea5af0aaf7aa13 |
| institution | Directory Open Access Journal |
| issn | 2199-160X |
| language | English |
| last_indexed | 2024-03-11T21:25:37Z |
| publishDate | 2023-06-01 |
| publisher | Wiley-VCH |
| record_format | Article |
| series | Advanced Electronic Materials |
| spelling | doaj.art-1e1f0d6a7df54916a6ea5af0aaf7aa132023-09-28T04:47:42ZengWiley-VCHAdvanced Electronic Materials2199-160X2023-06-0196n/an/a10.1002/aelm.202300018Configurable NbOx Memristors as Artificial Synapses or Neurons Achieved by Regulating the Forming Compliance Current for the Spiking Neural NetworkChuan Yu Han0Sheng Li Fang1Yi Lin Cui2Weihua Liu3Shi Quan Fan4Xiao Dong Huang5Xin Li6Xiao Li Wang7Guo He Zhang8Wing Man Tang9P. T. Lai10Jia Liu11Xianjie Wan12Zhou Yu13Li Geng14School of Microelectronics Xi'an Jiaotong University Xi'an 710049 P. R. ChinaSchool of Microelectronics Xi'an Jiaotong University Xi'an 710049 P. R. ChinaSchool of Microelectronics Xi'an Jiaotong University Xi'an 710049 P. R. ChinaSchool of Microelectronics Xi'an Jiaotong University Xi'an 710049 P. R. ChinaSchool of Microelectronics Xi'an Jiaotong University Xi'an 710049 P. R. ChinaKey Laboratory of MEMS of the Ministry of Education School of Electronic Science and Engineering Southeast University Nanjing 211189 ChinaSchool of Microelectronics Xi'an Jiaotong University Xi'an 710049 P. R. ChinaSchool of Microelectronics Xi'an Jiaotong University Xi'an 710049 P. R. ChinaSchool of Microelectronics Xi'an Jiaotong University Xi'an 710049 P. R. ChinaDepartment of Electrical and Electronic Engineering The University of Hong Kong Hong Kong 999077 ChinaDepartment of Electrical and Electronic Engineering The University of Hong Kong Hong Kong 999077 ChinaNo.24 Institute China Electronics Technology Group Corporation Chongqing 404100 P. R. ChinaNo.24 Institute China Electronics Technology Group Corporation Chongqing 404100 P. R. ChinaNo.24 Institute China Electronics Technology Group Corporation Chongqing 404100 P. R. ChinaSchool of Microelectronics Xi'an Jiaotong University Xi'an 710049 P. R. ChinaAbstract For the first time, a configurable NbOx memristor is achieved that can be configured as an artificial synapse or neuron after fabrication by controlling the forming compliance current (FCC). When the FCC ≤ 2 mA, the memristors exhibit the resistive‐switching (RS) property, enabling multiple types of synaptic plasticity, including short‐term potentiation, paired‐pulse facilitation, short‐term memory, and long‐term memory. When the FCC ≥ 3 mA, the memristors can be electroformed and exhibit the threshold switching (TS) property with excellent endurance (>1012), thus achieving various biological neuron characteristics, such as threshold‐triggering, strength‐modulation of spike frequency, and leaky integrate‐and‐fire. This enables the successful implementation of a spiking Pavlov's dog that employs the spikes as information carrier by connecting an RS NbOx memristor as artificial synapse and a TS memristor as artificial neuron in series. Furthermore, a fully NbOx memristors‐based single‐layer spiking neural network is simulated. It is first found that, due to the forgetting property of synapse, the recognition accuracy for the Modified National Institute of Standards and Technology handwritten digits is increased from 85.49% to 91.45%. This study provides a solid foundation for the development of neuromorphic machines based on the principles of the human brain.https://doi.org/10.1002/aelm.202300018artificial neuronsartificial synapsesforming compliance currentNbO x memristorsspiking neural networks |
| spellingShingle | Chuan Yu Han Sheng Li Fang Yi Lin Cui Weihua Liu Shi Quan Fan Xiao Dong Huang Xin Li Xiao Li Wang Guo He Zhang Wing Man Tang P. T. Lai Jia Liu Xianjie Wan Zhou Yu Li Geng Configurable NbOx Memristors as Artificial Synapses or Neurons Achieved by Regulating the Forming Compliance Current for the Spiking Neural Network Advanced Electronic Materials artificial neurons artificial synapses forming compliance current NbO x memristors spiking neural networks |
| title | Configurable NbOx Memristors as Artificial Synapses or Neurons Achieved by Regulating the Forming Compliance Current for the Spiking Neural Network |
| title_full | Configurable NbOx Memristors as Artificial Synapses or Neurons Achieved by Regulating the Forming Compliance Current for the Spiking Neural Network |
| title_fullStr | Configurable NbOx Memristors as Artificial Synapses or Neurons Achieved by Regulating the Forming Compliance Current for the Spiking Neural Network |
| title_full_unstemmed | Configurable NbOx Memristors as Artificial Synapses or Neurons Achieved by Regulating the Forming Compliance Current for the Spiking Neural Network |
| title_short | Configurable NbOx Memristors as Artificial Synapses or Neurons Achieved by Regulating the Forming Compliance Current for the Spiking Neural Network |
| title_sort | configurable nbox memristors as artificial synapses or neurons achieved by regulating the forming compliance current for the spiking neural network |
| topic | artificial neurons artificial synapses forming compliance current NbO x memristors spiking neural networks |
| url | https://doi.org/10.1002/aelm.202300018 |
| work_keys_str_mv | AT chuanyuhan configurablenboxmemristorsasartificialsynapsesorneuronsachievedbyregulatingtheformingcompliancecurrentforthespikingneuralnetwork AT shenglifang configurablenboxmemristorsasartificialsynapsesorneuronsachievedbyregulatingtheformingcompliancecurrentforthespikingneuralnetwork AT yilincui configurablenboxmemristorsasartificialsynapsesorneuronsachievedbyregulatingtheformingcompliancecurrentforthespikingneuralnetwork AT weihualiu configurablenboxmemristorsasartificialsynapsesorneuronsachievedbyregulatingtheformingcompliancecurrentforthespikingneuralnetwork AT shiquanfan configurablenboxmemristorsasartificialsynapsesorneuronsachievedbyregulatingtheformingcompliancecurrentforthespikingneuralnetwork AT xiaodonghuang configurablenboxmemristorsasartificialsynapsesorneuronsachievedbyregulatingtheformingcompliancecurrentforthespikingneuralnetwork AT xinli configurablenboxmemristorsasartificialsynapsesorneuronsachievedbyregulatingtheformingcompliancecurrentforthespikingneuralnetwork AT xiaoliwang configurablenboxmemristorsasartificialsynapsesorneuronsachievedbyregulatingtheformingcompliancecurrentforthespikingneuralnetwork AT guohezhang configurablenboxmemristorsasartificialsynapsesorneuronsachievedbyregulatingtheformingcompliancecurrentforthespikingneuralnetwork AT wingmantang configurablenboxmemristorsasartificialsynapsesorneuronsachievedbyregulatingtheformingcompliancecurrentforthespikingneuralnetwork AT ptlai configurablenboxmemristorsasartificialsynapsesorneuronsachievedbyregulatingtheformingcompliancecurrentforthespikingneuralnetwork AT jialiu configurablenboxmemristorsasartificialsynapsesorneuronsachievedbyregulatingtheformingcompliancecurrentforthespikingneuralnetwork AT xianjiewan configurablenboxmemristorsasartificialsynapsesorneuronsachievedbyregulatingtheformingcompliancecurrentforthespikingneuralnetwork AT zhouyu configurablenboxmemristorsasartificialsynapsesorneuronsachievedbyregulatingtheformingcompliancecurrentforthespikingneuralnetwork AT ligeng configurablenboxmemristorsasartificialsynapsesorneuronsachievedbyregulatingtheformingcompliancecurrentforthespikingneuralnetwork |