Abstract

Photonic integrated circuits are facilitating the development of optical neural networks, which have the potential to be both faster and more energy efficient than their electronic counterparts since optical signals are especially well-suited for implementing matrix multiplications. However, accurate programming of photonic chips for optical matrix multiplication remains a difficult challenge. Here, we describe both simple analytical models and data-driven models for offline training of optical matrix multipliers. We train and evaluate the models using experimental data obtained from a fabricated chip featuring a Mach-Zehnder interferometer mesh implementing 3-by-3 matrix multiplication. The neural network-based models outperform the simple physics-based models in terms of prediction error. Furthermore, the neural network models are also able to predict the spectral variations in the matrix weights for up to 100 frequency channels covering the C-band. The use of neural network models for programming the chip for optical matrix multiplication yields increased performance on multiple machine learning tasks.

Addressing data scarcity in optical matrix multiplier modeling using transfer learning

Ali Cem, Ognjen Jovanovic, Siqi Yan, Yunhong Ding, Darko Zibar, and Francesco Da Ros
Opt. Lett. 48(24) 6553-6556 (2023)

In situ optical backpropagation training of diffractive optical neural networks

Tiankuang Zhou, Lu Fang, Tao Yan, Jiamin Wu, Yipeng Li, Jingtao Fan, Huaqiang Wu, Xing Lin, and Qionghai Dai
Photon. Res. 8(6) 940-953 (2020)

Interferometry technique to eliminate noise in interference data based on a Mach–Zehnder interferometer

Amin Azari, Ezedin Mohajerani, Zahra Abedi, and Afshin Shams
Appl. Opt. 53(9) 1734-1738 (2014)

Single Mach–Zehnder interferometer based optical Boolean logic gates

Cláudia Reis, Tanay Chattopadhyay, Paulo André, and António Teixeira
Appl. Opt. 51(36) 8693-8701 (2012)

Electro-optic mode switch based on lithium-niobate Mach–Zehnder interferometer

Mengruo Zhang, Kaixin Chen, Wei Jin, and Kin Seng Chiang
Appl. Opt. 55(16) 4418-4422 (2016)