• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Applied Computational Electromagnetics Society journal >Artificial Synapse with Mnemonic Functionality using GSST-based Photonic Integrated Memory
【24h】

Artificial Synapse with Mnemonic Functionality using GSST-based Photonic Integrated Memory

机译：具有基于GSST的光子集成存储器的助记功能的人工突触

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Here we present a multi-level discrete-state nonvolatilephotonic memory based on an ultra-compact (<4μm)hybrid phase change material GSST-silicon Mach Zehndermodulator, with low insertion losses (3dB), to serve as node in aphotonic neural network. Emulating an opportunely trained100 × 100 fully connected multilayered perceptron neuralnetwork with this weighting functionality embedded as photonicmemory, shows up to 92% inference accuracy and robustnesstowards noise when performing predictions of unseen data.
机译：在这里，我们提出了一个多级离散状态的非易失性 光子内存基于超紧凑（<4μm） 杂交相变材料GSST-Silicon Mach Zehnder 调制器，具有低插入损耗（3DB），用作节点 光子神经网络。 模仿机会训练 100×100完全连接的多层的默认 网络具有嵌入到光子的此加权功能 内存，显示出高达92％的推理准确性和鲁棒性 在执行看不见数据的预测时致噪声。

著录项

  • 来源
  • 作者

    Mario Miscuglio; Jiawei Meng; Armin Mehrabian; Volker J. Sorger; Omer Yesiliurt; Ludmila J. Prokopeva; Alexander V. Kildishev; Yifei Zhang; Juejun Hu;

  • 作者单位

    Department of Electrical and ComputerEngineeringGeorge Washington UniversityWashington DC 20052 USA;

    Department of Electrical and ComputerEngineeringGeorge Washington UniversityWashington DC 20052 USA;

    Department of Electrical and ComputerEngineeringGeorge Washington UniversityWashington DC 20052 USA;

    Department of Electrical and ComputerEngineeringGeorge Washington UniversityWashington DC 20052 USA;

    Birck Nanotechnology CenterSchool of ECE Purdue UniversityWest Lafayette IN 47907 USA;

    Birck Nanotechnology CenterSchool of ECE Purdue UniversityWest Lafayette IN 47907 USA;

    Birck Nanotechnology CenterSchool of ECE Purdue UniversityWest Lafayette IN 47907 USA;

    Department of Materials Science &EngineeringMassachusetts Institute of TechnologyCambridge MA USA;

    Department of Materials Science &EngineeringMassachusetts Institute of TechnologyCambridge MA USA;

  • 收录信息
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

    GSST; phase change memories; photonic memories;

    机译：GSST;相变记忆;光子回忆;

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号