Simultaneous identification of models and parameters of scientific simulators

arXiv:2305.15174v2 Announce Type: replace
Abstract: Many scientific models are composed of multiple discrete components, and scientists often make heuristic decisions about which components to include. Bayesian inference provides a mathematical framework for systematically selecting model components, but defining prior distributions over model components and developing associated inference schemes has been challenging. We approach this problem in a simulation-based inference framework: We define model priors over candidate components and, from model simulations, train neural networks to infer joint probability distributions over …

abstract arxiv bayesian bayesian inference components cs.lg decisions framework identification inference multiple parameters prior scientific scientists type