Very High-Energy GRB Afterglow Fits with Neural-Network Emulation and Bayesian Inference
My rating
This paper targets the astrophysical modeling of very high-energy gamma-ray burst afterglow emission in the synchrotron self-Compton regime. Neural-network emulation is used to accelerate calculations involving time-dependent coupled particle-photon kinetic equations and Klein-Nishina suppression. Bayesian inference is applied to parameter estimation when fully numerical modeling is computationally expensive for techniques such as MCMC sampling.
Tags: