Statistical Unbounded Verification

Applicant Professor Jan Kretinsky, Ph.D.
Subject Area Theoretical Computer Science
Term from 2017 to 2022
Project identifier Deutsche Forschungsgemeinschaft (DFG) - Project number 383882557
 

Project Description

This project will develop formal methods for verification of probabilistic systems using simulations. In particular, it will provide algorithms for statistical model checking of large and/or unknown Markov chains and Markov decision processes against various objectives with generally unbounded horizon, a theoretically challenging and practically relevant problem. The main focus lies on linear temporal logic, mean payoff and linear distances.
DFG Programme Research Grants