Maximizing the information learned from finite data selects a simple model [Physics]

We use the language of uninformative Bayesian prior choice to study the selection of appropriately simple effective models. We advocate for the prior which maximizes the mutual information between parameters and predictions, learning as much as possible from limited data. When many parameters are poorly constrained by the available data,...

Source: Proceedings of the National Academy of Sciences - February 20, 2018 Category: Science Authors: Henry H. Mattingly, Mark K. Transtrum, Michael C. Abbott, Benjamin B. Machta Tags: Physical Sciences Source Type: research