Large-Scale Integrated Photonic Device Platform for Energy-Efficient AI/ML Accelerators

Bassem Tossoun; Xian Xiao; Stanley Cheung; Yuan Yuan; Yiwei Peng; Sudharsanan Srinivasan; George Giamougiannis; Zhihong Huang; Prerana Singaraju; Yanir London; Matej Hejda; Sri Priya Sundararajan; Yingtao Hu; Zheng Gong; Jongseo Baek; Antoine Descos; Morten Kapusta; Fabian Bohm; Thomas Van Vaerenbergh; Marco Fiorentino; Geza Kurczveil; Di Liang; Raymond G. Beausoleil

doi:10.1109/JSTQE.2025.3527904

Large-Scale Integrated Photonic Device Platform for Energy-Efficient AI/ML Accelerators

Bassem Tossoun, Xian Xiao, Stanley Cheung, Yuan Yuan, Yiwei Peng, Sudharsanan Srinivasan, George Giamougiannis, Zhihong Huang, Prerana Singaraju, Yanir London, Matej Hejda, Sri Priya Sundararajan, Yingtao Hu, Zheng Gong, Jongseo Baek, Antoine Descos, Morten Kapusta, Fabian Bohm, Thomas Van Vaerenbergh, Marco FiorentinoGeza Kurczveil, Di Liang, Raymond G. Beausoleil

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

Abstract

The convergence of deep learning and big data has spurred significant interest in developing novel hardware that can run large artificial intelligence (AI) workloads more efficiently. Over the last several years, silicon photonics has emerged as a disruptive technology for next-generation accelerators for machine learning (ML). More recently, the heterogeneous integration of III-V compound semiconductors has opened the door to integrating lasers and semiconductor optical amplifiers at wafer-scale, enabling the scaling of the size, density, and complexity of silicon photonic integrated circuits (PICs). Furthermore, using this technology, all of the individual components required to execute the operations within a neural network are available and can be integrated on the same PIC. Here, we review our innovations of an energy-efficient and scalable silicon photonic platform serving as the underlying foundation for next-generation AI accelerator hardware.

Original language	English
Journal	IEEE Journal of Selected Topics in Quantum Electronics
DOIs	https://doi.org/10.1109/JSTQE.2025.3527904
State	Accepted/In press - 1 Jan 2025
Externally published	Yes

Keywords

heterogeneous integration
memristors
neuromorphic computing
optical computing
silicon photonics

ASJC Scopus subject areas

Atomic and Molecular Physics, and Optics
Electrical and Electronic Engineering

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/JSTQE.2025.3527904

Cite this

Tossoun, B., Xiao, X., Cheung, S., Yuan, Y., Peng, Y., Srinivasan, S., Giamougiannis, G., Huang, Z., Singaraju, P., London, Y., Hejda, M., Sundararajan, S. P., Hu, Y., Gong, Z., Baek, J., Descos, A., Kapusta, M., Bohm, F., Van Vaerenbergh, T., ... Beausoleil, R. G. (Accepted/In press). Large-Scale Integrated Photonic Device Platform for Energy-Efficient AI/ML Accelerators. IEEE Journal of Selected Topics in Quantum Electronics. https://doi.org/10.1109/JSTQE.2025.3527904

@article{662121f52fa44184b4de56d3b52f3778,

title = "Large-Scale Integrated Photonic Device Platform for Energy-Efficient AI/ML Accelerators",

abstract = "The convergence of deep learning and big data has spurred significant interest in developing novel hardware that can run large artificial intelligence (AI) workloads more efficiently. Over the last several years, silicon photonics has emerged as a disruptive technology for next-generation accelerators for machine learning (ML). More recently, the heterogeneous integration of III-V compound semiconductors has opened the door to integrating lasers and semiconductor optical amplifiers at wafer-scale, enabling the scaling of the size, density, and complexity of silicon photonic integrated circuits (PICs). Furthermore, using this technology, all of the individual components required to execute the operations within a neural network are available and can be integrated on the same PIC. Here, we review our innovations of an energy-efficient and scalable silicon photonic platform serving as the underlying foundation for next-generation AI accelerator hardware.",

keywords = "heterogeneous integration, memristors, neuromorphic computing, optical computing, silicon photonics",

author = "Bassem Tossoun and Xian Xiao and Stanley Cheung and Yuan Yuan and Yiwei Peng and Sudharsanan Srinivasan and George Giamougiannis and Zhihong Huang and Prerana Singaraju and Yanir London and Matej Hejda and Sundararajan, {Sri Priya} and Yingtao Hu and Zheng Gong and Jongseo Baek and Antoine Descos and Morten Kapusta and Fabian Bohm and {Van Vaerenbergh}, Thomas and Marco Fiorentino and Geza Kurczveil and Di Liang and Beausoleil, {Raymond G.}",

note = "Publisher Copyright: {\textcopyright} 1995-2012 IEEE.",

year = "2025",

month = jan,

day = "1",

doi = "10.1109/JSTQE.2025.3527904",

language = "English",

journal = "IEEE Journal of Selected Topics in Quantum Electronics",

issn = "1077-260X",

publisher = "Institute of Electrical and Electronics Engineers",

}

Tossoun, B, Xiao, X, Cheung, S, Yuan, Y, Peng, Y, Srinivasan, S, Giamougiannis, G, Huang, Z, Singaraju, P, London, Y, Hejda, M, Sundararajan, SP, Hu, Y, Gong, Z, Baek, J, Descos, A, Kapusta, M, Bohm, F, Van Vaerenbergh, T, Fiorentino, M, Kurczveil, G, Liang, D & Beausoleil, RG 2025, 'Large-Scale Integrated Photonic Device Platform for Energy-Efficient AI/ML Accelerators', IEEE Journal of Selected Topics in Quantum Electronics. https://doi.org/10.1109/JSTQE.2025.3527904

TY - JOUR

T1 - Large-Scale Integrated Photonic Device Platform for Energy-Efficient AI/ML Accelerators

AU - Tossoun, Bassem

AU - Xiao, Xian

AU - Cheung, Stanley

AU - Yuan, Yuan

AU - Peng, Yiwei

AU - Srinivasan, Sudharsanan

AU - Giamougiannis, George

AU - Huang, Zhihong

AU - Singaraju, Prerana

AU - London, Yanir

AU - Hejda, Matej

AU - Sundararajan, Sri Priya

AU - Hu, Yingtao

AU - Gong, Zheng

AU - Baek, Jongseo

AU - Descos, Antoine

AU - Kapusta, Morten

AU - Bohm, Fabian

AU - Van Vaerenbergh, Thomas

AU - Fiorentino, Marco

AU - Kurczveil, Geza

AU - Liang, Di

AU - Beausoleil, Raymond G.

PY - 2025/1/1

Y1 - 2025/1/1

N2 - The convergence of deep learning and big data has spurred significant interest in developing novel hardware that can run large artificial intelligence (AI) workloads more efficiently. Over the last several years, silicon photonics has emerged as a disruptive technology for next-generation accelerators for machine learning (ML). More recently, the heterogeneous integration of III-V compound semiconductors has opened the door to integrating lasers and semiconductor optical amplifiers at wafer-scale, enabling the scaling of the size, density, and complexity of silicon photonic integrated circuits (PICs). Furthermore, using this technology, all of the individual components required to execute the operations within a neural network are available and can be integrated on the same PIC. Here, we review our innovations of an energy-efficient and scalable silicon photonic platform serving as the underlying foundation for next-generation AI accelerator hardware.

AB - The convergence of deep learning and big data has spurred significant interest in developing novel hardware that can run large artificial intelligence (AI) workloads more efficiently. Over the last several years, silicon photonics has emerged as a disruptive technology for next-generation accelerators for machine learning (ML). More recently, the heterogeneous integration of III-V compound semiconductors has opened the door to integrating lasers and semiconductor optical amplifiers at wafer-scale, enabling the scaling of the size, density, and complexity of silicon photonic integrated circuits (PICs). Furthermore, using this technology, all of the individual components required to execute the operations within a neural network are available and can be integrated on the same PIC. Here, we review our innovations of an energy-efficient and scalable silicon photonic platform serving as the underlying foundation for next-generation AI accelerator hardware.

KW - heterogeneous integration

KW - memristors

KW - neuromorphic computing

KW - optical computing

KW - silicon photonics

UR - http://www.scopus.com/inward/record.url?scp=85214847690&partnerID=8YFLogxK

U2 - 10.1109/JSTQE.2025.3527904

DO - 10.1109/JSTQE.2025.3527904

M3 - Article

AN - SCOPUS:85214847690

SN - 1077-260X

JO - IEEE Journal of Selected Topics in Quantum Electronics

JF - IEEE Journal of Selected Topics in Quantum Electronics

ER -

Large-Scale Integrated Photonic Device Platform for Energy-Efficient AI/ML Accelerators

Abstract

Keywords

ASJC Scopus subject areas

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this