Wafer-Scale Dynamic Neural-Network-System with Optical Waveguide

具有光波导的晶圆级动态神经网络系统

• 批准号: 12305024
• 负责人: KOYANAGI Mitsumasa
• 金额: $ 28.28万
• 依托单位: Tohoku University
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (A)
• 财政年份: 2000
• 资助国家: 日本
• 起止时间: 2000 至 2002
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-12305024/
• 关键词: Optical Waveguide; Parallel Processing; Wafer Level Integration; System on Chip; Wafer Bonding Technology; Micro-Bump; Semiconductor Technology; Large Scale Integration Technology; ウエーハレベルインテグレーション; ウエーハ張り合わせ技術; システムオンチップ

We proposed a new dynamic neural network system with optical wave guides. Living nerves have several thousand connections, namely synapses. These connections play an important role in living information processing system and provide recognition and association for human. The optical wave guide can simultaneously sum up many input signals like synapses. This study aimed in study the possibility of such optical neural network system with optical wave guides and to fabricate the test chips and test modules for the system. We adopted the learning algorithm using simultaneous perturbation for the dynamic neural network system with optical wave guides. We can simultaneously update the weights of synapses by the same increment in the learning algorithm using simultaneous perturbation. Therefore, the learning algorithm using simultaneous perturbation is very suitable for the dynamic neural network system with optical wave guides. We proposed new synapse circuit and learning circuit for this system. We also proposed a dynamic neural network system with three-dimensionally stacked structure. The re configurable circuits are used in this system in addition to the optical waveguides for synaptic connection. In order in fabricate the test chips and teat modules for this system, basic technologies such as the optical interconnection technology using waveguides and the three-dimensional integration technology using re-configurable circuits were developed.

我们提出了一种新的光波导动态神经网络系统。活着的神经有几千个连接，即突触。这些连接在生活信息处理系统中扮演着重要的角色，并为人类提供识别和关联。这种光波导能同时汇总许多像突触一样的输入信号。本研究的目的是研究这种具有光波导的光神经网络系统的可能性，并制作用于该系统的测试芯片和测试模块。对于具有光波导的动态神经网络系统，我们采用了同时摄动学习算法。在使用同步扰动的学习算法中，我们可以通过相同的增量同时更新突触的权重。因此，同时摄动学习算法非常适用于具有光波导的动态神经网络系统。我们为该系统提出了新的突触电路和学习电路。我们还提出了一种三维堆叠结构的动态神经网络系统。在该系统中，除了用于突触连接的光波导之外，还使用了可重新配置的电路。为了制作该系统的测试芯片和测试模块，开发了基于光波导的光互连技术和基于可重构电路的三维集成技术等基础技术。

项目成果

小柳光正, 栗野浩之: "VLSIの光インターコネクション"応用物理学会 関西セミナー SFM講演会予稿集. 55-55 (2002)

Mitsumasa Koyanagi、Hiroyuki Kurino：“VLSI 光互连”日本应用物理学会关西研讨会 SFM 讲座论文集 55-55 (2002)。

T.Hashimoto K.Kasahara,K,T,Park H.Kurino and M.Koyanagi: "Shared Memory with Optical Interconnection for Parallele Processor System"SPIE2001. (2001)

T.Hashimoto K.Kasahara、K,T、Park H.Kurino 和 M.Koyanagi：“用于并行处理器系统的具有光学互连的共享存储器”SPIE2001。

Y.Nakagawa,K.W.Lee,H.Kurino,M.Koyanagi and et al.: "Neuromorphic Analog Circuits for Three-Dimensional Stacked Vision Chip"Proc.of 7th International Conference on Neural Information Processing. 636-641 (2000)

Y.Nakakawa、K.W.Lee、H.Kurino、M.Koyanagi 等人：“三维堆叠视觉芯片的神经形态模拟电路”第七届国际神经信息处理会议论文集。

Tomonori NAKAMURA, Toru HASHIMOTO, Mitsumasa KOYANAGI, et al.: "Optical MCM technology for 3D LSI"Technical Report of IEICE. 2002-08. 55-60 (2002)

Tomonori NAKAMURA、Toru HASHIMOTO、Mitsumasa KOYANAGI 等：“3D LSI 的光学 MCM 技术”IEICE 技术报告。

Hoon Choi, Hyuckjae Oh, Hiroyuki Kurino, Mitsumasa Koyanagi, et al.: "SiGe Selective Epitaxial Growth on SOI Substrate for The Elevated Source and Drain Structure"Abstracts of International Semiconductor Technology Conference (ISTC). 55-55 (2002)

Hoon Choi、Hyuckjae Oh、Hiroyuki Kurino、Mitsumasa Koyanagi 等人：“SiGe Selective Epitcentric Growth on SOI Substrate for The Elevated Source and Drain Structure”国际半导体技术会议（ISTC）摘要。