ProbLog2: From probabilistic programming to statistical relational learning

摘要

ProbLog is a probabilistic programming language based on Prolog. The new ProbLog system - called ProbLog2 - can solve a range of inference and learning tasks typical for the Probabilistic Graphical Models (PGM) and Statistical Relational Learning (SRL) communities. The main mechanism behind ProbLog2 is a conversion of the given program to a weighted Boolean formula. We argue that this conversion approach can also be applied - with certain restrictions - to other probabilistic programming languages such as Church and Figaro.