A Verified Compiler for Probability Density Functions


Title: A Verified Compiler for Probability Density Functions
Authors: Manuel Eberl, Johannes Hölzl and Tobias Nipkow
Submission date: 2014-10-09
Abstract: Bhat et al. [TACAS 2013] developed an inductive compiler that computes density functions for probability spaces described by programs in a probabilistic functional language. In this work, we implement such a compiler for a modified version of this language within the theorem prover Isabelle and give a formal proof of its soundness w.r.t. the semantics of the source and target language. Together with Isabelle's code generation for inductive predicates, this yields a fully verified, executable density compiler. The proof is done in two steps: First, an abstract compiler working with abstract functions modelled directly in the theorem prover's logic is defined and proved sound. Then, this compiler is refined to a concrete version that returns a target-language expression.

An article with the same title and authors is published in the proceedings of ESOP 2015. A detailed presentation of this work can be found in the first author's master's thesis.

  author  = {Manuel Eberl and Johannes Hölzl and Tobias Nipkow},
  title   = {A Verified Compiler for Probability Density Functions},
  journal = {Archive of Formal Proofs},
  month   = oct,
  year    = 2014,
  note    = {\url{http://isa-afp.org/entries/Density_Compiler.shtml},
            Formal proof development},
  ISSN    = {2150-914x},
License: BSD License