Evolution of Bio‐Inspired Artificial Synapses: Materials, Structures, and Mechanisms,Small

当前位置： X-MOL 学术 › Small › 论文详情

Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)

Evolution of Bio‐Inspired Artificial Synapses: Materials, Structures, and Mechanisms
Small ( IF 13.0 ) Pub Date : 2020-05-26 , DOI: 10.1002/smll.202000041
Haiyang Yu ₁ , Huanhuan Wei ₁ , Jiangdong Gong ₁ , Hong Han ₁ , Mingxue Ma ₁ , Yongfei Wang ₂ , Wentao Xu ₁

Affiliation

Artificial synapses (ASs) are electronic devices emulating important functions of biological synapses, which are essential building blocks of artificial neuromorphic networks for brain‐inspired computing. A human brain consists of several quadrillion synapses for information storage and processing, and massively parallel computation. Neuromorphic systems require ASs to mimic biological synaptic functions, such as paired‐pulse facilitation, short‐term potentiation, long‐term potentiation, spatiotemporally‐correlated signal processing, and spike‐timing‐dependent plasticity, etc. Feature size and energy consumption of ASs need to be minimized for high‐density energy‐efficient integration. This work reviews recent progress on ASs. First, synaptic plasticity and functional emulation are introduced, and then synaptic electronic devices for neuromorphic computing systems are discussed. Recent advances in flexible artificial synapses for artificial sensory nerves are also briefly introduced. Finally, challenges and opportunities in the field are discussed.

中文翻译：

受生物启发的人工突触的进化：材料，结构和机制

人工突触（ASs）是模仿生物突触重要功能的电子设备，它们是人工神经形态网络对于大脑启发式计算的重要组成部分。人脑由几个四维突触组成，用于信息存储和处理以及大规模并行计算。神经形态系统要求AS模仿生物突触功能，例如成对脉冲促进，短期增强，长期增强，时空相关的信号处理以及与尖峰时序相关的可塑性等。AS的特征尺寸和能量消耗需要最小化以实现高密度节能集成。这项工作回顾了AS的最新进展。首先，介绍突触可塑性和功能仿真，然后讨论了用于神经形态计算系统的突触电子设备。还简要介绍了用于人工感觉神经的柔性人工突触的最新进展。最后，讨论了该领域的挑战和机遇。

更新日期：2020-05-26

点击分享查看原文

点击收藏

阅读更多本刊新发论文本刊介绍/投稿指南