Predicting the turbulent transport of cosmic rays via neural networks

TL;DR

This paper introduces a neural network model that rapidly predicts cosmic ray transport in turbulent magnetic fields, achieving high accuracy and significantly reducing computational time compared to traditional simulations.

Contribution

The authors develop a neural network-based method for fast and accurate prediction of cosmic ray transport coefficients in turbulent astrophysical environments.

Findings

Neural network predicts transport coefficients 10^7 times faster than traditional simulations.

Achieves an overall prediction error of approximately 5%.

Demonstrates effectiveness on synthetic datasets generated from test-particle simulations.

Abstract

A fast artificial neural network is developed for the prediction of cosmic ray transport in turbulent astrophysical magnetic fields. The setup is trained and tested on bespoke datasets that are constructed with the aid of test-particle numerical simulations of relativistic cosmic ray dynamics in synthetic stochastic fields. The neural network uses, as input, particle and field properties and estimates transport coefficients 10^7 faster than standard numerical simulations with an overall error of ~5% .