Applications of deep learning to relativistic hydrodynamics

TL;DR

This paper reviews recent progress in applying deep learning, specifically a stacked U-net neural network, to model and predict the complex non-linear evolution in relativistic hydrodynamics efficiently.

Contribution

It introduces a novel deep neural network architecture that accurately captures relativistic hydrodynamics and rapidly predicts final profiles from initial conditions.

Findings

The stacked U-net effectively models non-linear hydrodynamic evolution.

The neural network can quickly predict final profiles for various initial conditions.

Deep learning offers a promising tool for relativistic hydrodynamics simulations.

Abstract

In this proceeding, we will briefly review our recent progress on implementing deep learning to relativistic hydrodynamics. We will demonstrate that a successfully designed and trained deep neural network, called {\tt stacked U-net}, can capture the main features of the non-linear evolution of hydrodynamics, which could also rapidly predict the final profiles for various testing initial conditions.