Tags

Type your tag names separated by a space and hit enter

Bipolar Analog Memristors as Artificial Synapses for Neuromorphic Computing.
Materials (Basel) 2018; 11(11)M

Abstract

Synaptic devices with bipolar analog resistive switching behavior are the building blocks for memristor-based neuromorphic computing. In this work, a fully complementary metal-oxide semiconductor (CMOS)-compatible, forming-free, and non-filamentary memristive device (Pd/Al₂O₃/TaOx/Ta) with bipolar analog switching behavior is reported as an artificial synapse for neuromorphic computing. Synaptic functions, including long-term potentiation/depression, paired-pulse facilitation (PPF), and spike-timing-dependent plasticity (STDP), are implemented based on this device; the switching energy is around 50 pJ per spike. Furthermore, for applications in artificial neural networks (ANN), determined target conductance states with little deviation (<1%) can be obtained with random initial states. However, the device shows non-linear conductance change characteristics, and a nearly linear conductance change behavior is obtained by optimizing the training scheme. Based on these results, the device is a promising emulator for biology synapses, which could be of great benefit to memristor-based neuromorphic computing.

Authors+Show Affiliations

Institute of Microelectronics of Chinese Academy of Sciences, Beijing 100029, China. wangrui@ime.ac.cn. University of Chinese Academy of Sciences, Beijing 100049, China. wangrui@ime.ac.cn.Institute of Microelectronics of Chinese Academy of Sciences, Beijing 100029, China. shituo@ime.ac.cn. University of Chinese Academy of Sciences, Beijing 100049, China. shituo@ime.ac.cn.Institute of Microelectronics of Chinese Academy of Sciences, Beijing 100029, China. zhaoxiaolong@ime.ac.cn. University of Chinese Academy of Sciences, Beijing 100049, China. zhaoxiaolong@ime.ac.cn.Institute of Microelectronics of Chinese Academy of Sciences, Beijing 100029, China. wangwei_esss@nudt.edu.cn.Institute of Microelectronics of Chinese Academy of Sciences, Beijing 100029, China. weijinsong@ime.ac.cn. University of Science and Technology of China, Hefei 230026, China. weijinsong@ime.ac.cn.Institute of Microelectronics of Chinese Academy of Sciences, Beijing 100029, China. lujian@ime.ac.cn. University of Science and Technology of China, Hefei 230026, China. lujian@ime.ac.cn.Institute of Microelectronics of Chinese Academy of Sciences, Beijing 100029, China. zhaoxiaolong@ime.ac.cn.Institute of Microelectronics of Chinese Academy of Sciences, Beijing 100029, China. wuzuheng@ime.ac.cn. University of Chinese Academy of Sciences, Beijing 100049, China. wuzuheng@ime.ac.cn.Institute of Microelectronics of Chinese Academy of Sciences, Beijing 100029, China. caorongrong@ime.ac.cn. University of Chinese Academy of Sciences, Beijing 100049, China. caorongrong@ime.ac.cn.University of Science and Technology of China, Hefei 230026, China. longshibing@ime.ac.cn.Institute of Microelectronics of Chinese Academy of Sciences, Beijing 100029, China. liuqi@ime.ac.cn. University of Chinese Academy of Sciences, Beijing 100049, China. liuqi@ime.ac.cn.Institute of Microelectronics of Chinese Academy of Sciences, Beijing 100029, China. liuming@ime.ac.cn. University of Chinese Academy of Sciences, Beijing 100049, China. liuming@ime.ac.cn.

Pub Type(s)

Journal Article

Language

eng

PubMed ID

30373122

Citation

Wang, Rui, et al. "Bipolar Analog Memristors as Artificial Synapses for Neuromorphic Computing." Materials (Basel, Switzerland), vol. 11, no. 11, 2018.
Wang R, Shi T, Zhang X, et al. Bipolar Analog Memristors as Artificial Synapses for Neuromorphic Computing. Materials (Basel). 2018;11(11).
Wang, R., Shi, T., Zhang, X., Wang, W., Wei, J., Lu, J., ... Liu, M. (2018). Bipolar Analog Memristors as Artificial Synapses for Neuromorphic Computing. Materials (Basel, Switzerland), 11(11), doi:10.3390/ma11112102.
Wang R, et al. Bipolar Analog Memristors as Artificial Synapses for Neuromorphic Computing. Materials (Basel). 2018 Oct 26;11(11) PubMed PMID: 30373122.
* Article titles in AMA citation format should be in sentence-case
TY - JOUR T1 - Bipolar Analog Memristors as Artificial Synapses for Neuromorphic Computing. AU - Wang,Rui, AU - Shi,Tuo, AU - Zhang,Xumeng, AU - Wang,Wei, AU - Wei,Jinsong, AU - Lu,Jian, AU - Zhao,Xiaolong, AU - Wu,Zuheng, AU - Cao,Rongrong, AU - Long,Shibing, AU - Liu,Qi, AU - Liu,Ming, Y1 - 2018/10/26/ PY - 2018/09/10/received PY - 2018/10/18/revised PY - 2018/10/23/accepted PY - 2018/10/31/entrez PY - 2018/10/31/pubmed PY - 2018/10/31/medline KW - artificial synapse KW - memristor KW - neuromorphic computing JF - Materials (Basel, Switzerland) JO - Materials (Basel) VL - 11 IS - 11 N2 - Synaptic devices with bipolar analog resistive switching behavior are the building blocks for memristor-based neuromorphic computing. In this work, a fully complementary metal-oxide semiconductor (CMOS)-compatible, forming-free, and non-filamentary memristive device (Pd/Al₂O₃/TaOx/Ta) with bipolar analog switching behavior is reported as an artificial synapse for neuromorphic computing. Synaptic functions, including long-term potentiation/depression, paired-pulse facilitation (PPF), and spike-timing-dependent plasticity (STDP), are implemented based on this device; the switching energy is around 50 pJ per spike. Furthermore, for applications in artificial neural networks (ANN), determined target conductance states with little deviation (<1%) can be obtained with random initial states. However, the device shows non-linear conductance change characteristics, and a nearly linear conductance change behavior is obtained by optimizing the training scheme. Based on these results, the device is a promising emulator for biology synapses, which could be of great benefit to memristor-based neuromorphic computing. SN - 1996-1944 UR - https://www.unboundmedicine.com/medline/citation/30373122/Bipolar_Analog_Memristors_as_Artificial_Synapses_for_Neuromorphic_Computing_ L2 - http://www.mdpi.com/resolver?pii=ma11112102 DB - PRIME DP - Unbound Medicine ER -