Quantitative Evaluation of Systems, International Conference on (2005)

Torino, Italy

Sept. 19, 2005 to Sept. 22, 2005

ISBN: 0-7695-2427-3

pp: 155-165

DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/QEST.2005.8

Kousha Etessami , University of Edinburgh

Mihalis Yannakakis , Columbia University

ABSTRACT

We present algorithms for the qualitative and quantitative model checking of Linear Temporal Logic (LTL) properties for Recursive Markov Chains (RMCs). Recursive Markov Chains are a natural abstract model of procedural probabilistic programs and related systems involving recursion and probability. For the qualitative problem ("Given a RMC A and an LTL formula \wp, do the computations of A satisfy \wp almost surely?") we present an algorithm that runs in polynomial space in A and exponential time in \wp. For several classes of RMCs, including RMCs with one exit (a special case that corresponds to well-studied probabilistic systems, e.g., multi-type branching processes and stochastic context-free grammars) the algorithm runs in polynomial time in A and exponential time in \wp. On the other hand, we also prove that the problem is EXPTIMEhard, and hence it is EXPTIME-complete. For the quantitative problem ("does the probability that a computation of A satisfies \wp exceed a given threshold p?", or approximate the probability within a desired precision) we present an algorithm that runs in polynomial space in A and exponential space in \wp. For linearly-recursive RMCs, we can compute the exact probability in time polynomial in A and exponential in \wp. These results improve by one exponential, in both the qualitative and quantitative case, the complexity that one would obtain if one first translated the LTL formula to a Buchi automaton and then applied the model checking algorithm for Buchi automata from [11]. Our results combine techniques developed in [10, 11] for analysis of RMCs, and in [6] for LTL model checking of flat Markov Chains

INDEX TERMS

null

CITATION

Kousha Etessami,
Mihalis Yannakakis,
"Checking LTL Properties of Recursive Markov Chains",

*Quantitative Evaluation of Systems, International Conference on*, vol. 00, no. , pp. 155-165, 2005, doi:10.1109/QEST.2005.8