Limits of nonlinear and dispersive fiber propagation for an optical fiber-based extreme learning machine

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ermolaev, Andrei V., Hary, Mathilde, Leybov, Lev, Ryczkowski, Piotr, Skalli, Anas, Brunner, Daniel, Genty, Goëry, Dudley, John M.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912423777140736
author Ermolaev, Andrei V.
Hary, Mathilde
Leybov, Lev
Ryczkowski, Piotr
Skalli, Anas
Brunner, Daniel
Genty, Goëry
Dudley, John M.
author_facet Ermolaev, Andrei V.
Hary, Mathilde
Leybov, Lev
Ryczkowski, Piotr
Skalli, Anas
Brunner, Daniel
Genty, Goëry
Dudley, John M.
contents We report a generalized nonlinear Schrödinger equation simulation model of an extreme learning machine (ELM) based on optical fiber propagation. Using the MNIST handwritten digit dataset as a benchmark, we study how accuracy depends on propagation dynamics, as well as parameters governing spectral encoding, readout, and noise. For this dataset and with quantum noise limited input, test accuracies of : over 91% and 93% are found for propagation in the anomalous and normal dispersion regimes respectively. Our results also suggest that quantum noise on the input pulses introduces an intrinsic penalty to ELM performance.
format Preprint
id arxiv_https___arxiv_org_abs_2503_03649
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Limits of nonlinear and dispersive fiber propagation for an optical fiber-based extreme learning machine
Ermolaev, Andrei V.
Hary, Mathilde
Leybov, Lev
Ryczkowski, Piotr
Skalli, Anas
Brunner, Daniel
Genty, Goëry
Dudley, John M.
Optics
Machine Learning
We report a generalized nonlinear Schrödinger equation simulation model of an extreme learning machine (ELM) based on optical fiber propagation. Using the MNIST handwritten digit dataset as a benchmark, we study how accuracy depends on propagation dynamics, as well as parameters governing spectral encoding, readout, and noise. For this dataset and with quantum noise limited input, test accuracies of : over 91% and 93% are found for propagation in the anomalous and normal dispersion regimes respectively. Our results also suggest that quantum noise on the input pulses introduces an intrinsic penalty to ELM performance.
title Limits of nonlinear and dispersive fiber propagation for an optical fiber-based extreme learning machine
topic Optics
Machine Learning
url https://arxiv.org/abs/2503.03649