An internal shock model calibrated with real gamma-ray burst light curves using a genetic algorithm

Abstract

The origin of gamma-ray burst (GRB) prompt emission remains an open question. The internal shock (IS) model is a leading scenario for converting relativistic ejecta kinetic energy into gamma rays, but its parameters have not yet been fully calibrated against observed GRB light curves (LCs) to reproduce their diversity. We adopt a machine-learning framework to optimise the IS model by comparing simulated and observed LC properties from three GRB catalogues (Swift/BAT, Fermi/GBM, CGRO/BATSE). Assu…