TY - JOUR
T1 - Photonic pattern reconstruction enabled by on-chip online learning and inference
AU - Marquez, Bicky A.
AU - Guo, Zhimu
AU - Morison, Hugh
AU - Shekhar, Sudip
AU - Chrostowski, Lukas
AU - Prucnal, Paul
AU - Shastri, Bhavin J.
N1 - Funding Information:
This work was supported in part by the Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery Grants Program and the Collaborative Research and Development (CRD) Grant with Huawei Canada.
Publisher Copyright:
© 2021 The Author(s). Published by IOP Publishing Ltd
PY - 2021/4
Y1 - 2021/4
N2 - Recent investigations in neuromorphic photonics exploit optical device physics for neuron models, and optical interconnects for distributed, parallel, and analog processing. Integrated solutions enabled by silicon photonics enable high-bandwidth, low-latency and low switching energy, making it a promising candidate for special-purpose artificial intelligence hardware accelerators. Here, we experimentally demonstrate a silicon photonic chip that can perform training and testing of a Hopfield network, i.e. recurrent neural network, via vector dot products. We demonstrate that after online training, our trained Hopfield network can successfully reconstruct corrupted input patterns.
AB - Recent investigations in neuromorphic photonics exploit optical device physics for neuron models, and optical interconnects for distributed, parallel, and analog processing. Integrated solutions enabled by silicon photonics enable high-bandwidth, low-latency and low switching energy, making it a promising candidate for special-purpose artificial intelligence hardware accelerators. Here, we experimentally demonstrate a silicon photonic chip that can perform training and testing of a Hopfield network, i.e. recurrent neural network, via vector dot products. We demonstrate that after online training, our trained Hopfield network can successfully reconstruct corrupted input patterns.
KW - Artificial intelligence hardware
KW - Brain-inspired computing
KW - Neuromorphic photonics
KW - Photonic integrated circuits
KW - Recurrent neural network
UR - http://www.scopus.com/inward/record.url?scp=85102480779&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85102480779&partnerID=8YFLogxK
U2 - 10.1088/2515-7647/abe3d9
DO - 10.1088/2515-7647/abe3d9
M3 - Article
AN - SCOPUS:85102480779
SN - 2515-7647
VL - 3
JO - JPhys Photonics
JF - JPhys Photonics
IS - 2
M1 - 024006
ER -