Computing by Means of Physics-Based Optical Neural Networks

TL;DR

This paper explores the use of physics-based opto-electronic hardware to perform biology-inspired neural network computations, highlighting high-speed capabilities and current challenges in practical implementation.

Contribution

It introduces a novel approach combining biological neural models with physics-based optical hardware for high-speed computation.

Findings

Potential for very-high-speed computation

Identification of key technical challenges

Call for interdisciplinary solutions

Abstract

We report recent research on computing with biology-based neural network models by means of physics-based opto-electronic hardware. New technology provides opportunities for very-high-speed computation and uncovers problems obstructing the wide-spread use of this new capability. The Computation Modeling community may be able to offer solutions to these cross-boundary research problems.