Quantified naturalness from Bayesian statistics

Abstract

We present a formulation of naturalness made in the framework of Bayesian statistics, which unravels the conceptual problems related to previous approaches. Among other things, the relative interpretation of the measure of naturalness turns out to be unambiguously established by Jeffreys' scale. Also, the usual sensitivity formulation (so-called Barbieri-Giudice measure) appears to be embedded in our formulation under an extended form. We derive the general sensitivity formula applicable to an arbitrary number of observables. Several consequences and developments are further discussed. As a final illustration, we work out the map of combined fine-tuning associated to the gauge hierarchy problem and neutralino dark matter in a classic supersymmetric model.