Probabilistic Verification and Approximation

  • Authors:

  • Richard Lassaigne;Sylvain Peyronnet

  • Affiliations:

  • Equipe de Logique Mathmatique, UMR 7056 CNRS, University Paris VII;EPITA Research and Development Laboratory (LRDE)

  • Venue:
  • Electronic Notes in Theoretical Computer Science (ENTCS)
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

Model checking is an algorithmic method allowing to automatically verify if a system which is represented as a Kripke model satisfies a given specification. Specifications are usually expressed by formulas of temporal logic. The first objective of this paper is to give an overview of methods able to verify probabilistic systems. Models of such systems are labelled discrete time Markov chains and specifications are expressed in extensions of temporal logic by probabilistic operators. The second objective is to focus on the complexity of these methods and to answer the question: can probabilistic verification be efficiently approximated? In general, the answer is negative. However, in many applications, the specification formulas can be expressed in some positive fragment of linear time temporal logic. In this paper, we show how some simple randomized approximation algorithms can improve the efficiency of the verification of such probabilistic specifications.