ProbPoly
Călin-Rareş Turliuc · 2011
We propose a novel framework for Probabilistic Inductive Logic Programming (PILP), called ProbPoly, in the context of Stochastic Logic Programs (SLP). The approach aims at learning probabilities of new learned clauses by minimizing the mean squared error (MSE) between the observed example probabilities and the predicted probabilities. We illustrate the approach using a very simple Probabilistic Context-Fee Grammar (PCFG). We then introduce a basic method for revising the probabilities of a simple discrete time Markov chain (DTMC) using an integration of ProbPoly and a probabilistic model checker, so that the properties, which were initially violated, are satisfied in the new DTMC. This work provides the foundations for a general approach to requirements elaboration in probabilistic settings.