Phase space topology of four-wave mixing reconstructed by a neural network

TL;DR

This paper demonstrates how a neural network can reconstruct the phase space topology of four-wave mixing in optical fibers, accurately predicting nonlinear dynamics and key features like recurrence cycles and separatrix boundaries.

Contribution

It introduces a machine learning approach to reconstruct complex optical fiber dynamics from experimental spectral data, revealing detailed phase space structures.

Findings

Neural network accurately predicts dynamics over tens of kilometers.

Main features of phase space topology are successfully retrieved.

Multiple Fermi-Pasta-Ulam recurrence cycles identified.

Abstract

The dynamics of ideal four-wave mixing in optical fiber is reconstructed by taking advantage of the combination of experimental measurements with supervised machine learning strategies. The training data consist of power-dependent spectral phase and amplitude recorded at the output of a short segment of fiber. The neural network is able to accurately predict the nonlinear dynamics over tens of kilometers, and to retrieve the main features of the phase space topology including multiple Fermi-Pasta-Ulam recurrence cycles and the system separatrix boundary.