Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group
Conferences > OFC > 2020 > W3A > W3A.5

All-optical recurrent neural network with sigmoid activation function

George Mourgias-Alexandris, George Dabos, Nikolaos Passalis, Anastasios Tefas, Angelina Totovic, and Nikos Pleros

Not Accessible

Your library or personal account may give you access

Abstract

We demonstrate experimentally, the first all-optical recurrent-neuron with a sigmoid activation function and four WDM-inputs with 100psec pulses. The proposed neuron geared up a neural-network for financial prediction-tasks exhibiting an accuracy of 42.57% on FI-2010.

© 2020 The Author(s)

PDF Article
More Like This
Adaptive all-optical sigmoid activation functions for Photonic Neural Networks using Fabry-Perot laser diodes under optical injection

Petar Atanasijević, Christos Pappas, Mladen Banović, Jasna Crnjanski, Apostolos Tsakyridis, Miltiadis Moralis-Pegios, Konstantinos Vyrsokinos, Marko Krstić, Peđa Mihailović, Slobodan Petričević, Nikos Pleros, and Dejan Gvozdić
Tu3F.2 Optical Fiber Communication Conference (OFC) 2024

Experimental demonstration of an optical neuron with a logistic Sigmoid activation function

George Mourgias-Alexandris, Apostolos Tsakyridis, Nikolaos Passalis, Anastasios Tefas, and Nikos Pleros
Th2A.37 Optical Fiber Communication Conference (OFC) 2019

Photonic Recurrent Neural Networks with Gating Circuit

George Dabos, George Mourgias-Alexandris, Angelina Totovic, Nikolaos Passalis, Anastasios Tefas, and Nikos Pleros
JTU2C.8 CLEO: Applications and Technology (CLEO:A&T) 2020

View More...
Select as filters
Select Topics Cancel
Top
       
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.
Privacy | Terms of Use