Optical activation function using a metamaterial waveguide for an all-optical neural network

TL;DR

This paper demonstrates an enhanced nonlinear optical activation function using a metamaterial-integrated silicon waveguide, achieving high inference accuracy and paving the way for power-efficient all-optical neural networks.

Contribution

It introduces a metamaterial structure to significantly boost the nonlinear optical coefficient of silicon waveguides for neural network applications.

Findings

Enhanced two-photon absorption coefficient by 1200 times.

Achieved 98.36% accuracy in handwritten character recognition.

Demonstrated potential for power-efficient all-optical neural networks.

Abstract

In this study, we experimentally demonstrated that the nonlinear optical coefficient of the original Si can be enhanced by incorporating a metamaterial structure into an existing silicon waveguide. The two-photon absorption coefficient enhanced by the metamaterial structure was 424 cm/GW, which is 1.2x10^3 times higher than that of Si. Using this metamaterial waveguide-based nonlinear optical activation function, we achieved a high inference accuracy of 98.36% in the handwritten character recognition task, comparable to that obtained with the ReLU function as the activation function. Therefore, our approach can contribute to the realization of more power-efficient and compact all-optical neural networks.