Limitations of opto-electronic neural networks

Jeffrey Yu; Alan Johnston; Demetri Psaltis; David Brady

doi:10.1117/12.951514

Limitations of opto-electronic neural networks

Jeffrey Yu, Alan Johnston, Demetri Psaltis, David Brady

Research output: Contribution to journal › Article › peer-review

Abstract

Neural networks are composed of three components: Neurons, weights, and connections. Limitation on implementing these components in electronics and optics are discussed. We show that using electronically fabricated neurons and a combination of optics and electronics for the weights and connections efficiently utilizes the advantage of each technology. The choice of the technology used to implement the weights and connections depends on the type of neural network being built.

Original language	English (US)
Pages (from-to)	40-51
Number of pages	12
Journal	Proceedings of SPIE - The International Society for Optical Engineering
Volume	1053
DOIs	https://doi.org/10.1117/12.951514
State	Published - Jun 29 1989
Externally published	Yes

ASJC Scopus subject areas

Electronic, Optical and Magnetic Materials
Condensed Matter Physics
Computer Science Applications
Applied Mathematics
Electrical and Electronic Engineering

Access to Document

10.1117/12.951514

Cite this

@article{1acd9004c3c7402fbdffe5921168a045,

title = "Limitations of opto-electronic neural networks",

abstract = "Neural networks are composed of three components: Neurons, weights, and connections. Limitation on implementing these components in electronics and optics are discussed. We show that using electronically fabricated neurons and a combination of optics and electronics for the weights and connections efficiently utilizes the advantage of each technology. The choice of the technology used to implement the weights and connections depends on the type of neural network being built.",

author = "Jeffrey Yu and Alan Johnston and Demetri Psaltis and David Brady",

note = "Funding Information: At Caltech, this research is funded by DARPA and the Air Force Office of Scientific Research. At JPL, this research is funded by DARPA.",

year = "1989",

month = jun,

day = "29",

doi = "10.1117/12.951514",

language = "English (US)",

volume = "1053",

pages = "40--51",

journal = "Proceedings of SPIE - The International Society for Optical Engineering",

issn = "0277-786X",

publisher = "SPIE",

}

TY - JOUR

T1 - Limitations of opto-electronic neural networks

AU - Yu, Jeffrey

AU - Johnston, Alan

AU - Psaltis, Demetri

AU - Brady, David

N1 - Funding Information: At Caltech, this research is funded by DARPA and the Air Force Office of Scientific Research. At JPL, this research is funded by DARPA.

PY - 1989/6/29

Y1 - 1989/6/29

N2 - Neural networks are composed of three components: Neurons, weights, and connections. Limitation on implementing these components in electronics and optics are discussed. We show that using electronically fabricated neurons and a combination of optics and electronics for the weights and connections efficiently utilizes the advantage of each technology. The choice of the technology used to implement the weights and connections depends on the type of neural network being built.

AB - Neural networks are composed of three components: Neurons, weights, and connections. Limitation on implementing these components in electronics and optics are discussed. We show that using electronically fabricated neurons and a combination of optics and electronics for the weights and connections efficiently utilizes the advantage of each technology. The choice of the technology used to implement the weights and connections depends on the type of neural network being built.

UR - http://www.scopus.com/inward/record.url?scp=84958490015&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84958490015&partnerID=8YFLogxK

U2 - 10.1117/12.951514

DO - 10.1117/12.951514

M3 - Article

AN - SCOPUS:84958490015

SN - 0277-786X

VL - 1053

SP - 40

EP - 51

JO - Proceedings of SPIE - The International Society for Optical Engineering

JF - Proceedings of SPIE - The International Society for Optical Engineering

ER -

Limitations of opto-electronic neural networks

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this