Erlangen 2026 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

GR: Fachverband Gravitation, Relativistische Astrophysik und Kosmologie

GR 19: Numerical Relativity III

GR 19.4: Talk

Friday, March 20, 2026, 11:45–12:00, KH 02.012

Machine Learning-Accelerated HLLD Riemann Solver for GRMHD — •Keneth Miler — Institut für Theoretische Physik, Goethe Universität, Max-von-Laue-Str. 1, D-60438

We present a machine-learning-enhanced HLLD Riemann solver for GRMHD simulations that significantly reduces computational cost. The primary bottleneck in HLLD schemes is the iterative pressure recovery from conserved variables. We replace this expensive root-finding procedure with a trained neural network that directly predicts primitive pressure.

Keywords: GRMHD; Riemann solver; Machine learning