期刊
NPG ASIA MATERIALS
卷 10, 期 -, 页码 1097-1106出版社
NATURE PUBLISHING GROUP
DOI: 10.1038/s41427-018-0101-y
关键词
-
资金
- National Research Foundation of Korea [NRF-2016R1C1B2007330, NRF-2018R1A2B6008104]
- KU-KIST research fund
- Samsung Electronics
- Korea University Future Research Grant
- KIST Institutional Program [2V05750]
- National Research Foundation of Korea [2016R1C1B2007330] Funding Source: Korea Institute of Science & Technology Information (KISTI), National Science & Technology Information Service (NTIS)
The human brain intrinsically operates with a large number of synapses, more than 10(15). Therefore, one of the most critical requirements for constructing artificial neural networks (ANNs) is to achieve extremely dense synaptic array devices, for which the crossbar architecture containing an artificial synaptic node at each cross is indispensable. However, crossbar arrays suffer from the undesired leakage of signals through neighboring cells, which is a major challenge for implementing ANNs. In this work, we show that this challenge can be overcome by using Pt/TaOy/nanoporous (NP) TaOx/Ta memristor synapses because of their self-rectifying behavior, which is capable of suppressing unwanted leakage pathways. Moreover, our synaptic device exhibits high non-linearity (up to 10(4)), low synapse coupling (S.C, up to 4.00 x 10(-5)), acceptable endurance (5000 cycles at 85 degrees C), sweeping (1000 sweeps), retention stability and acceptable cell uniformity. We also demonstrated essential synaptic functions, such as long-term potentiation (LTP), long-term depression (LTD), and spiking-timing-dependent plasticity (STDP), and simulated the recognition accuracy depending on the S.C for MNIST handwritten digit images. Based on the average S.C (1.60 x 10(-4)) in the fabricated crossbar array, we confirmed that our memristive synapse was able to achieve an 89.08% recognition accuracy after only 15 training epochs.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据