TY - JOUR
T1 - Primer on silicon neuromorphic photonic processors
T2 - Architecture and compiler
AU - Ferreira De Lima, Thomas
AU - Tait, Alexander N.
AU - Mehrabian, Armin
AU - Nahmias, Mitchell A.
AU - Huang, Chaoran
AU - Peng, Hsuan Tung
AU - Marquez, Bicky A.
AU - Miscuglio, Mario
AU - El-Ghazawi, Tarek
AU - Sorger, Volker J.
AU - Shastri, Bhavin J.
AU - Prucnal, Paul R.
N1 - Funding Information:
This article is supported by National Science Foundation (NSF) (Award numbers 1740262, 1740235, 1642991); SRC nCore (Award number 2018-C-A); Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery Grants Program; Canadian Foundation of Innovation (CFI) John R. Evans Fund (JELF); Ontario Research Fund: Small Infrastructure Program.
Funding Information:
Funding: This article is supported by National Science Foundation (NSF) (Award numbers 1740262, 1740235, 1642991); SRC nCore (Award number 2018-NC-2763-A); Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery Grants Program; Canadian Foundation of Innovation (CFI) John R. Evans Fund (JELF); Ontario Research Fund: Small Infrastructure Program.
Publisher Copyright:
© 2020 Thomas Ferreira de Lima et al., published by De Gruyter, Berlin/Boston 2020.
PY - 2020/10/1
Y1 - 2020/10/1
N2 - Microelectronic computers have encountered challenges in meeting all of today's demands for information processing. Meeting these demands will require the development of unconventional computers employing alternative processing models and new device physics. Neural network models have come to dominate modern machine learning algorithms, and specialized electronic hardware has been developed to implement them more efficiently. A silicon photonic integration industry promises to bring manufacturing ecosystems normally reserved for microelectronics to photonics. Photonic devices have already found simple analog signal processing niches where electronics cannot provide sufficient bandwidth and reconfigurability. In order to solve more complex information processing problems, they will have to adopt a processing model that generalizes and scales. Neuromorphic photonics aims to map physical models of optoelectronic systems to abstract models of neural networks. It represents a new opportunity for machine information processing on sub-nanosecond timescales, with application to mathematical programming, intelligent radio frequency signal processing, and real-time control. The strategy of neuromorphic engineering is to externalize the risk of developing computational theory alongside hardware. The strategy of remaining compatible with silicon photonics externalizes the risk of platform development. In this perspective article, we provide a rationale for a neuromorphic photonics processor, envisioning its architecture and a compiler. We also discuss how it can be interfaced with a general purpose computer, i.e. a CPU, as a coprocessor to target specific applications. This paper is intended for a wide audience and provides a roadmap for expanding research in the direction of transforming neuromorphic photonics into a viable and useful candidate for accelerating neuromorphic computing.
AB - Microelectronic computers have encountered challenges in meeting all of today's demands for information processing. Meeting these demands will require the development of unconventional computers employing alternative processing models and new device physics. Neural network models have come to dominate modern machine learning algorithms, and specialized electronic hardware has been developed to implement them more efficiently. A silicon photonic integration industry promises to bring manufacturing ecosystems normally reserved for microelectronics to photonics. Photonic devices have already found simple analog signal processing niches where electronics cannot provide sufficient bandwidth and reconfigurability. In order to solve more complex information processing problems, they will have to adopt a processing model that generalizes and scales. Neuromorphic photonics aims to map physical models of optoelectronic systems to abstract models of neural networks. It represents a new opportunity for machine information processing on sub-nanosecond timescales, with application to mathematical programming, intelligent radio frequency signal processing, and real-time control. The strategy of neuromorphic engineering is to externalize the risk of developing computational theory alongside hardware. The strategy of remaining compatible with silicon photonics externalizes the risk of platform development. In this perspective article, we provide a rationale for a neuromorphic photonics processor, envisioning its architecture and a compiler. We also discuss how it can be interfaced with a general purpose computer, i.e. a CPU, as a coprocessor to target specific applications. This paper is intended for a wide audience and provides a roadmap for expanding research in the direction of transforming neuromorphic photonics into a viable and useful candidate for accelerating neuromorphic computing.
KW - Neuromorphic computing
KW - Optical neural networks
KW - Photonic integrated circuits
KW - Silicon photonics
KW - Ultrafast information processing
UR - http://www.scopus.com/inward/record.url?scp=85093642808&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85093642808&partnerID=8YFLogxK
U2 - 10.1515/nanoph-2020-0172
DO - 10.1515/nanoph-2020-0172
M3 - Review article
AN - SCOPUS:85093642808
SN - 2192-8606
VL - 9
SP - 4055
EP - 4073
JO - Nanophotonics
JF - Nanophotonics
IS - 13
ER -