Likelihood-free inference: unlocking the full potential of simulations

Likelihood-free inference is an emerging technology that promises tighter constraints on physical parameters than traditional methods when an analytic form of the likelihood is intractable and simulations are employed. We will introduce likelihood-free inference (aka simulation-based inference) from low-level principles and discuss where neural inference ought to outperform traditional methods. We will also explore several recent, promising results that tackle outstanding, practical hurdles to neural inference in particle physics, before finally surveying the work still to be done to make this a day-to-day reality in the physical sciences.