Refuting Equivalence in Probabilistic Programs with Conditioning
Open Access
- 1 May 2025
- book chapter
- Published by Springer Nature
- p. 279-300
Abstract
We consider the problem of refuting equivalence of probabilistic programs, i.e., the problem of proving that two probabilistic programs induce different output distributions. We study this problem in the context of programs with conditioning (i.e., with observe and score statements), where the output distribution is conditioned by the event that all the observe statements along a run evaluate to true, and where the probability densities of different runs may be updated via the score statements. Building on a recent work on programs without conditioning, we present a new equivalence refutation method for programs with conditioning. Our method is based on weighted restarting, a novel transformation of probabilistic programs with conditioning to the output equivalent probabilistic programs without conditioning that we introduce in this work. Our method is the first to be both a) fully automated, and b) providing provably correct answers. We demonstrate the applicability of our method on a set of programs from the probabilistic inference literature.Keywords
All Related Versions
This publication has 55 references indexed in Scilit:
- Probabilistic Program Verification via Inductive Synthesis of Inductive Invariants2023
- Does a Program Yield the Right Distribution?2022
- Sound and Complete Certificates for Quantitative Termination Analysis of Probabilistic Programs2022
- On Lexicographic Proof Rules for Probabilistic Termination2021
- Correctness by Construction for Probabilistic Programs2020
- Tail Probabilities for Randomized Program Runtimes via Martingales for Higher Moments2019
- PSense: Automatic Sensitivity Analysis for Probabilistic Programs2018
- Termination Analysis of Probabilistic Programs Through Positivstellensatz’s2016
- Understanding Probabilistic Programs2015
- On Probabilistic Program Equivalence and Refinement2005