Behavioral Plasticity Emulated with Lithium Lanthanum Titanate‐Based Memristive Devices: Habituation,Advanced Electronic Materials

当前位置： X-MOL 学术 › Adv. Electron. 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.)

Behavioral Plasticity Emulated with Lithium Lanthanum Titanate‐Based Memristive Devices: Habituation
Advanced Electronic Materials ( IF 5.3 ) Pub Date : 2017-07-17 , DOI: 10.1002/aelm.201700046
Tuo Shi ₁ , Jian-Fang Wu ₁ , Yong Liu ₂ , Rui Yang ₁ , Xin Guo ₁

Affiliation

Behavioral plasticity builds up an association between information and an appropriate behavioral response, enabling organisms to adjust their behaviors in response to stimulations from external environments. Habituation is the elementary and ubiquitous form of the behavioral plasticity. A direct emulation of the behavioral habituation with electronic devices is critical in neuromorphic computing. Herein, memristive devices based on a mixed conductor of lithium ions and electrons, lithium lanthanum titanate (LLTO), are fabricated in a Pt/LLTO/Pt structure. Because the physical processes of lithium‐ion migration and polaron hopping in LLTO are similar to the two neuromorphic processes proposed in the dual‐process theory of habituation, the Pt/LLTO/Pt memristive device enables a direct emulation of the habituation in the behavioral plasticity. More importantly, the nine parametric definitions in neuroscience for habituation are realized. The successful emulation of behavioral habituation represents an advance in the hardware implementation of neuromorphic functionalities and may simplify the structure of memristor‐based artificial neural networks.

中文翻译：

基于钛酸锂镧系忆阻器件的行为可塑性：习性

行为可塑性在信息和适当的行为反应之间建立起联系，使生物能够响应外部环境的刺激来调整其行为。习惯是行为可塑性的基本形式和普遍形式。电子设备对行为习惯的直接仿真在神经形态计算中至关重要。在此，以Pt / LLTO / Pt结构制造基于锂离子和电子的混合导体的忆阻器件，即钛酸锂镧（LLTO）。由于LLTO中锂离子迁移和极化子跳跃的物理过程与习惯性双过程理论中提出的两个神经形态过程相似，因此Pt / LLTO / Pt忆阻器件可以在行为可塑性上直接模拟习惯性。更重要的是，实现了神经科学中习惯性的九个参数定义。行为习惯的成功仿真代表了神经形态功能的硬件实现方面的进步，并可能简化基于忆阻器的人工神经网络的结构。

更新日期：2017-07-17

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