Optical neural network can process information in parallel by using the technology based on free-space and integrated platform. Over the last half century, the development of integrated circuits has been limited by Moore's law. We know that neural network is based on the digital computer for successive calculation, most of which cannot be made into real-time processing system. Therefore, it is necessary to develop ONN for real-time processing and device miniaturization. In this paper, we review the progress of optical neural networks. Firstly, based on the principle of artificial neural networks, we elaborate the essence of optical matrix multiplier for linear operation. Then we introduce the optical neural network achieved by free-space optical interconnection and waveguide optical interconnection. Finally we talk about the nonlinearity in optical neural networks. With the gradual maturity of nanotechnology and the rapid advancement of silicon photonic integrated circuits, the progress of integrated photonic neural network has been promoted. Therefore, the construction of optical neural network on the future integrated photonic platform has potential application value.

/pdf/a-review-of-optical-neural-networks-1mfz647iif.pdf

A Review of Optical Neural Networks

Scientists in the UK have made a high-resolution optically addressed spatial light modulator (OASLM) based on zinc oxide (ZnO) nanoparticles. There is a strong demand for inexpensive OASLMs capable of producing high-resolution holograms over large active areas. The three researchers, who are all from the University of Cambridge, used a simple and inexpensive method to fabricate their OASLM — spin coating ZnO nanoparticles suspended in ethanol onto an indium–tin-oxide-coated glass substrate and then producing a photosensitive layer by low temperature (under 180 °C) annealing. The OASLM had a resolution of up to 825 lp mm−1 when operated at a voltage of below 6 V. The researchers attribute this performance to the use of a photoconductive film having a thin layer of semiconductor nanoparticles and a nematic liquid crystal with a relatively sharp threshold voltage.

/pdf/a-high-resolution-optically-addressed-spatial-light-5besucmk9m.pdf

A high-resolution optically addressed spatial light modulator based on ZnO nanoparticles

Recent advances in materials technology has led to the development of new and improved optically nonlinear materials that can be used for a wide variety of applications ranging from deep UV conversion for laser fusion experiments to fast optical switching devices in future telecommunication systems. The processibility that many of these new materials have to offer will allow devices to be designed for specific applications. Some recent developments in the search for new materials are described and the future role that some of these may play in the design of nonlinear optical devices is discussed.

New materials and their characterization for photonic device applications

Large-scale, highly integrated and low-power-consuming hardware is becoming progressively more important for realizing optical neural networks (ONNs) capable of advanced optical computing. Traditional experimental implementations need N2 units such as Mach-Zehnder interferometers (MZIs) for an input dimension N to realize typical computing operations (convolutions and matrix multiplication), resulting in limited scalability and consuming excessive power. Here, we propose the integrated diffractive optical network for implementing parallel Fourier transforms, convolution operations and application-specific optical computing using two ultracompact diffractive cells (Fourier transform operation) and only N MZIs. The footprint and energy consumption scales linearly with the input data dimension, instead of the quadratic scaling in the traditional ONN framework. A ~10-fold reduction in both footprint and energy consumption, as well as equal high accuracy with previous MZI-based ONNs was experimentally achieved for computations performed on the MNIST and Fashion-MNIST datasets. The integrated diffractive optical network (IDNN) chip demonstrates a promising avenue towards scalable and low-power-consumption optical computational chips for optical-artificial-intelligence. 

/pdf/space-efficient-optical-computing-with-an-integrated-chip-1otadlim.pdf

Space-efficient optical computing with an integrated chip diffractive neural network

The general problems and limitations of photoaddressed liquid crystal spatial light modulators (SLMs) are discussed, and experimental results are presented. A GaAs wafer photoaddressed nematic liquid crystal SLM is described in detail with experimental results. The edge enhancing or differentiating SLM is discussed, and the results of GaAs photoaddressing are reported. The preliminary experimental behavior of photoaddressed ferroelectric liquid crystal SLMs is described.

Photoaddressed liquid crystal spatial light modulators

A high-speed, coherent spatial light modulator is under development. It is based on the coupling of a silicon photodiode and electrooptic crystal. The design criteria and theoretical performance are presented. Optical quality of \lambda/20 rms at \lambda = 500 nm over a 1 in aperture has been demonstrated. Submicrosecond write and erase rates are predicted. Heat dissipation limits the predicted continuous rate to 2.5 kHz or higher, depending on required optical quality. Calculated phase MTF is 50 percent at 10 cycles/mm.

W. Anderson

Papers

High-speed spatial light modulator