Design of a Silicon-based Optical Neural Network
Authors
Danni Zhang, Pengfei Wang, Guangzhen Luo, Yu Bi, Ye Zhang, Junkai Yi, Yanmei Su, Yejin Zhang, Jiaoqing Pan
Corresponding Author
Jiaoqing Pan
Available Online December 2019.
- DOI
- 10.2991/mmsta-19.2019.39How to use a DOI?
- Keywords
- optical neural network; simulation; SiN; chip design; neuromorphic computing
- Abstract
In this paper, a silicon-based fully connected optical neural network (ONN) is designed, which can be use to image classification and recognition with accuracy greater than 97% . A fully connected neural network is constructed. One layer model has been used in chip design. Chip simulation shows accuracy could not be impacted as photons large enough. This structure could be used in deep learning.
- Copyright
- © 2019, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - Danni Zhang AU - Pengfei Wang AU - Guangzhen Luo AU - Yu Bi AU - Ye Zhang AU - Junkai Yi AU - Yanmei Su AU - Yejin Zhang AU - Jiaoqing Pan PY - 2019/12 DA - 2019/12 TI - Design of a Silicon-based Optical Neural Network BT - Proceedings of the 2019 2nd International Conference on Mathematics, Modeling and Simulation Technologies and Applications (MMSTA 2019) PB - Atlantis Press SP - 184 EP - 186 SN - 2352-538X UR - https://doi.org/10.2991/mmsta-19.2019.39 DO - 10.2991/mmsta-19.2019.39 ID - Zhang2019/12 ER -