TechRxiv
arxiv.pdf (371.87 kB)
0/0

Photonic convolutional neural networks using integrated diffractive optics

Download (371.87 kB)
preprint
posted on 23.03.2020 by Jun Rong Ong, Thomas Yong Long Ang, Chin Chun Ooi, Soon Thor Lim, Ching Eng Png
With recent rapid advances in photonic integrated circuits, it has been demonstrated that programmable photonic chips can be used to implement artificial neural networks. Convolutional neural networks (CNN) are a class of deep learning methods that have been highly successful in applications such as image classification and speech processing. We present an architecture to implement a photonic CNN using the Fourier transform property of integrated star couplers. We show, in computer simulation, high accuracy image classification using the MNIST dataset. We also model component imperfections in photonic CNN and show that the performance degradation can be recovered in a programmable chip. Our proposed architecture provides a large reduction in physical footprint compared to current implementations as it utilizes the natural advantages of optics and hence offers a scalable pathway towards integrated photonic deep learning processors.

History

Email Address of Submitting Author

ongjr@ihpc.a-star.edu.sg

ORCID of Submitting Author

0000-0002-7488-3454

Submitting Author's Institution

Institute of High Performance Computing, A*STAR

Submitting Author's Country

Singapore

Licence

Exports

Read the peer-reviewed publication

in IEEE Journal of Selected Topics in Quantum Electronics

Licence

Exports