Nested sampling with any prior you like

ABSTRACT Nested sampling is an important tool for conducting Bayesian analysis in Astronomy and other fields, both complicated posterior distributions parameter inference, computing marginal likelihoods model comparison. One technical obstacle to using nested practice the requirement (for most common implementations) that prior be provided form of transformations from unit hyper-cube target density. For many applications – particularly when one experiment as another such a transformation not readily available. In this letter, we show parametric bijectors trained on samples desired density provide general purpose method constructing uniform base prior, enabling practical use under arbitrary priors. We demonstrate conjunction with number examples cosmology.