Electro-optical Neural Networks based on Time-stretch Method

09/13/2019
by   Yubin Zang, et al.
0

In this paper, a novel architecture of electro-optical neural networks based on the time-stretch method is proposed and numerically simulated. By stretching time-domain ultrashort pulses, multiplications of large scale weight matrices and vectors can be implemented on light and multiple-layer of feedforward neural network operations can be easily implemented with fiber loops. Via simulation, the performance of a three-layer electro-optical neural network is tested by the handwriting digit recognition task and the accuracy reaches 88 under considerable noise.

READ FULL TEXT

page 5

page 7

research
03/01/2017

Theoretical Properties for Neural Networks with Weight Matrices of Low Displacement Rank

Recently low displacement rank (LDR) matrices, or so-called structured m...
research
05/27/2021

Avancee-1 Mission and SaDoD Method: LiDAR-based stimulated atomic disintegration of space debris (SaDoD) using Optical Neural Networks

The surface degradation of satellites in Low Earth Orbit (LEO) is affect...
research
10/10/2022

Rejecting noise in Baikal-GVD data with neural networks

Baikal-GVD is a large (∼ 1 km^3) underwater neutrino telescope installed...
research
05/17/2019

Sequential training algorithm for neural networks

A sequential training method for large-scale feedforward neural networks...
research
04/24/2019

Optical machine learning with incoherent light and a single-pixel detector

The concept of optical diffractive neural network (DNN) is proposed rece...
research
09/01/2021

Acceleration Method for Learning Fine-Layered Optical Neural Networks

An optical neural network (ONN) is a promising system due to its high-sp...
research
12/13/2017

Throughput Scaling for MMF-Enabled Optical Datacenter Networks by Time-Slicing-Based Crosstalk Mitigation

Modal crosstalk is the main bottleneck in MMF-enabled optical datacenter...

Please sign up or login with your details

Forgot password? Click here to reset