Reconfigurable all-optical nonlinear activation functions for neuromorphic photonics

Abstract

We experimentally demonstrate all-optical reconfigurable nonlinear activation functions in a cavity-loaded Mach–Zehnder interferometer device on a silicon photonics platform, via the free-carrier dispersion effect. Our device is programmable to generate various nonlinear activation functions, including sigmoid, radial-basis, clamped rectified linear unit, and softplus, with tunable thresholds. We simulate benchmark tasks such as XOR and MNIST handwritten digit classifications with experimentally measured activation functions and obtain accuracies of 100% and 94%, respectively. Our device can serve as nonlinear units in photonic neural networks, while its nonlinear transfer function can be flexibly programmed to optimize the performance of different neuromorphic tasks.

Document Details

Document Type
Pub Defense Publication
Publication Date
Aug 26, 2020
Source ID
10.1364/ol.398234

Entities

People

  • Aashu Jha
  • Chaoran Huang
  • Paul Prucnal

Organizations

  • Office of Naval Research

Tags

Fields of Study

  • Physics

Readers

  • Integrated Circuit Design and Technology.
  • Neural Network Machine Learning.
  • Optical Physics and Photonics.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Machine Learning Algorithms
  • AI & ML - Neural Networks