Statistical model checking for parameterized models
Résumé
We propose a simulation-based technique, in the spirit of Statistical Model Checking, for approximate verification of probabilis-tic models with parametric transitions, and we focus in particular on parametric Markov chains. Our technique is based on an extension of Monte Carlo algorithms that allows to approximate the probability of satisfying a given finite trace property as a (polynomial) function of the parameters of the model. The confidence intervals associated with this approximation can also be expressed as a function of the parameters. In the paper, we present both the theoretical foundations of this technique and a prototype implementation in Python which we evaluate on a set of benchmarks.
Domaines
Modélisation et simulationOrigine | Fichiers produits par l'(les) auteur(s) |
---|
Loading...