Silicon Photonics Neural Networks for Training and Inference

Bhavin J. Shastri; Matthew J. Filipovich; Zhimu Guo; Paul R. Prucnal; Sudip Shekhar; Volker J. Sorger

Silicon Photonics Neural Networks for Training and Inference

Bhavin J. Shastri, Matthew J. Filipovich, Zhimu Guo, Paul R. Prucnal, Sudip Shekhar, Volker J. Sorger

Research output: Contribution to journal › Conference article › peer-review

Abstract

Deep learning hardware accelerators based on analog photonic networks are trained on standard digital electronics. We discuss on-chip training of neural networks enabled by a silicon photonic architecture for parallel, efficient, and fast data operations.

Original language	English (US)
Article number	NeW2D.2
Journal	Optics InfoBase Conference Papers
State	Published - 2022
Event	Photonic Networks and Devices, Networks 2022 - Maastricht, Netherlands Duration: Jul 24 2022 → Jul 28 2022

All Science Journal Classification (ASJC) codes

Electronic, Optical and Magnetic Materials
Mechanics of Materials

Cite this

@article{a291b74936f64932801f8d8a1f07434c,

title = "Silicon Photonics Neural Networks for Training and Inference",

abstract = "Deep learning hardware accelerators based on analog photonic networks are trained on standard digital electronics. We discuss on-chip training of neural networks enabled by a silicon photonic architecture for parallel, efficient, and fast data operations.",

author = "Shastri, \{Bhavin J.\} and Filipovich, \{Matthew J.\} and Zhimu Guo and Prucnal, \{Paul R.\} and Sudip Shekhar and Sorger, \{Volker J.\}",

note = "Publisher Copyright: {\textcopyright} 2022 The Author(s).; Photonic Networks and Devices, Networks 2022 ; Conference date: 24-07-2022 Through 28-07-2022",

year = "2022",

language = "English (US)",

journal = "Optics InfoBase Conference Papers",

issn = "2162-2701",

}

TY - JOUR

T1 - Silicon Photonics Neural Networks for Training and Inference

AU - Shastri, Bhavin J.

AU - Filipovich, Matthew J.

AU - Guo, Zhimu

AU - Prucnal, Paul R.

AU - Shekhar, Sudip

AU - Sorger, Volker J.

PY - 2022

Y1 - 2022

N2 - Deep learning hardware accelerators based on analog photonic networks are trained on standard digital electronics. We discuss on-chip training of neural networks enabled by a silicon photonic architecture for parallel, efficient, and fast data operations.

AB - Deep learning hardware accelerators based on analog photonic networks are trained on standard digital electronics. We discuss on-chip training of neural networks enabled by a silicon photonic architecture for parallel, efficient, and fast data operations.

UR - https://www.scopus.com/pages/publications/85146662858

UR - https://www.scopus.com/inward/citedby.url?scp=85146662858&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:85146662858

SN - 2162-2701

JO - Optics InfoBase Conference Papers

JF - Optics InfoBase Conference Papers

M1 - NeW2D.2

T2 - Photonic Networks and Devices, Networks 2022

Y2 - 24 July 2022 through 28 July 2022

ER -

Silicon Photonics Neural Networks for Training and Inference

Abstract

All Science Journal Classification (ASJC) codes

Other files and links

Fingerprint

Cite this