The Base Measure Problem and its Solution

10/06/2020
by   Alexey Radul, et al.
0

Probabilistic programming systems generally compute with probability density functions, leaving the base measure of each such function implicit. This mostly works, but creates problems when densities with respect to different base measures are accidentally combined or compared. Mistakes also happen when computing volume corrections for continuous changes of variables, which in general depend on the support measure. We motivate and clarify the problem in the context of a composable library of probability distributions and bijective transformations. We solve the problem by standardizing on Hausdorff measure as a base, and deriving formulas for comparing and combining mixed-dimension densities, as well as updating densities with respect to Hausdorff measure under diffeomorphic transformations. We also propose a software architecture that implements these formulas efficiently in the common case. We hope that by adopting our solution, probabilistic programming systems can become more robust and general, and make a broader class of models accessible to practitioners.

READ FULL TEXT
research
10/01/2021

Applied Measure Theory for Probabilistic Modeling

Probabilistic programming and statistical computing are vibrant areas in...
research
11/04/2022

Computing highest density regions for continuous univariate distributions with known probability functions

We examine the problem of computing the highest density region (HDR) in ...
research
01/26/2018

Concentration without measure

Although there doesn't exist the Lebesgue measure in the ball M of C[0,1...
research
02/17/2021

Dynamical behavior of alternate base expansions

We generalize the greedy and lazy β-transformations for a real base β to...
research
10/28/2022

Nonparametric Probabilistic Regression with Coarse Learners

Probabilistic Regression refers to predicting a full probability density...
research
01/26/2018

Average values of functionals and concentration without measure

Although there doesn't exist the Lebesgue measure in the ball M of C[0,1...
research
10/06/2021

Continuous logistic Gaussian random measure fields for spatial distributional modelling

We investigate a class of models for non-parametric estimation of probab...

Please sign up or login with your details

Forgot password? Click here to reset