Universal Linear Optics Revisited: New Perspectives for Neuromorphic Computing with Silicon Photonics

• Published in: IEEE journal of selected topics in quantum electronics Vol. 29; no. 2: Optical Computing; pp. 1 - 16
• Main Authors: Giamougiannis, George, Tsakyridis, Apostolos, Moralis-Pegios, Miltiadis, Totovic, Angelina R., Kirtas, Manos, Passalis, Nikolaos, Tefas, Anastasios, Lazovsky, David, Pleros, Nikos
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.03.2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Artificial Intelligence; Circuits; Deep Learning; High speed; High-speed optical techniques; Integrated photonics; Linear operators; Linear optics; Linear transformations; Mathematical analysis; Matrices (mathematics); Matrix decomposition; Multiport interferometers; Neural networks; Neuromorphic computing; Neuromorphic photonics; Optical attenuators; Optical interferometry; Optical losses; Optics; Photonics; Reprogrammable photonic meshes; Silicon photonics; State-of-the-art reviews; Switches
• ISSN: 1077-260X; 1558-4542
• DOI: 10.1109/JSTQE.2022.3228318

Reprogrammable optical meshes comprise a subject of heightened interest for the execution of linear transformations, having a significant impact in numerous applications that extend from the implementation of optical switches up to neuromorphic computing. Herein, we review the state-of-the-art approaches for the realization of unitary transformations and universal linear operators in the photonic domain and present our recent work in the field, that allows for fidelity restorable and low-loss optical circuitry with single-step programmability. These advantages unlock a new framework for matrix-vector multiplications required by neuromorphic silicon photonic circuits, supporting: i) high-speed and high-accuracy neural network (NN) inference, ii) high-speed tiled matrix multiplication, iii) NN training and iv) programmable photonic NNs. This new potential is initially validated through recent experimental results using SiGe EAM technology and static weights and, subsequently, utilized for demonstrating experimentally the first Deep NN (DNN) where optical tiled matrix multiplication up to 50GHz is realized, allowing optics to execute DNNs with large number of trainable parameters over a limited photonic hardware. Finally, the new performance framework is benchmarked against state-of-the-art NN processors and photonic NN roadmap projections, highlighting its perspectives to turn the energy and area efficiency promise of neuromorphic silicon photonics into a tangible reality.