High-Performance Neuromorphic Computing Based on Ferroelectric Synapses with Excellent Conductance Linearity and Symmetry,Advanced Functional Materials

当前位置： X-MOL 学术 › Adv. Funct. Mater. › 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.)

High-Performance Neuromorphic Computing Based on Ferroelectric Synapses with Excellent Conductance Linearity and Symmetry
Advanced Functional Materials ( IF 18.5 ) Pub Date : 2022-06-22 , DOI: 10.1002/adfm.202202366
Shu‐Ting Yang _{1,

2} , Xing‐Yu Li ₁ , Tong‐Liang Yu ₁ , Jie Wang _{3,

4} , Hong Fang _{3,

4} , Fang Nie ₁ , Bin He ₃ , Le Zhao ₂ , Wei‐Ming Lü _{3,

4} , Shi‐Shen Yan ₁ , Alain Nogaret ₅ , Gang Liu ₆ , Li‐Mei Zheng ₁

Affiliation

Artificial synapses can boost neuromorphic computing to overcome the inherent limitations of von Neumann architecture. As a promising memristor candidate, ferroelectric tunnel junctions (FTJ) enable the authors to successfully emulate spike-timing-dependent synapses. However, the nonlinear and asymmetric synaptic weight update under repeated presynaptic stimulation hampers neuromorphic computing by favoring the runaway of synaptic weights during learning. Here, the authors demonstrate an FTJ whose conductivity varies linearly and symmetrically by judiciously combining ferroelectric domain switching and oxygen vacancy migration. The artificial neural network based on this FTJ-synapse achieves classification accuracy of 96.7% during supervised learning, which is the closest to the maximum theoretical value of 98% achieved to date. This artificial synapse also demonstrates stable unsupervised learning in a noisy environment for its well-balanced spike-timing-dependent plasticity response. The novel concept of controlling ionic migration in ferroelectric materials paves the way toward highly reliable and reproducible supervised and unsupervised learning strategies.

中文翻译：

基于具有出色电导线性和对称性的铁电突触的高性能神经形态计算

人工突触可以增强神经形态计算，以克服冯诺依曼架构的固有局限性。作为有前途的忆阻器候选者，铁电隧道结 (FTJ) 使作者能够成功地模拟尖峰定时依赖的突触。然而，在重复突触前刺激下的非线性和不对称突触权重更新通过有利于学习期间突触权重的失控而阻碍了神经形态计算。在这里，作者通过明智地将铁电畴转换和氧空位迁移结合起来，展示了一种 FTJ，其电导率呈线性和对称变化。基于该 FTJ 突触的人工神经网络在监督学习期间实现了 96.7% 的分类准确率，这与迄今为止达到的 98% 的最大理论值最接近。这种人工突触还展示了在嘈杂环境中稳定的无监督学习，因为它具有均衡的尖峰时间依赖性可塑性响应。控制铁电材料中离子迁移的新概念为高度可靠和可重复的监督和无监督学习策略铺平了道路。

更新日期：2022-06-22

点击分享查看原文

点击收藏

阅读更多本刊最新论文本刊介绍/投稿指南11