Physics-Informed Neural Networks for the Relativistic Burgers Equation in the Exterior of a Schwarzschild Black Hole

Shuyang Xiang · 2025

Read Physics-Informed Neural Networks for the Relativistic Burgers Equation in the Exterior of a Schwarzschild Black Hole by Shuyang Xiang (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2506-00951.txt.

Description

Shuyang Xiang trains neural networks to solve a relativistic wave equation just outside a Schwarzschild black hole. His 2025 paper uses physics-informed neural networks, which learn by respecting the equation itself rather than only fitting data. The target is a relativistic Burgers equation, a model of flows that can form shock waves in curved spacetime.

The design adds a shock-aware network block and loss terms inspired by the Godunov scheme, a classic shock-capturing method. Tests with varied starting conditions showed the network reproducing both smooth solutions and sharp discontinuous ones. Xiang presents the work as a step toward machine solvers for equations in general relativity.

Catalog record

CREATORShuyang Xiang
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE18 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2506-00951

Browse the full library · About the Tower of Babel