A Testing Scenario for Probabilistic Automata

M.I.A. Stoelinga and F.W. Vaandrager. A Testing Scenario for Probabilistic Automata.
In J.C.M. Baeten, J.K. Lenstra, J. Parrow and G.J. Woeginger, editors, Proceedings ICALP'03, LNCS 2719, pages 407-418. Springer-Verlag, 2003.
This paper received the EATCS best paper award.
Full version available as L. Cheung, M.I.A. Stoelinga and F.W. Vaandrager. A Testing Scenario for Probabilistic Processes. Technical Report ICIS-R06002, ICIS, Radboud University Nijmegen, January 2006.

Abstract

We present a simple and intuitive testing scenario for image finite probabilistic automata. We introduce a notion of finite testing via a variant of the well-known machine. Under this scenario, two automata are deemed observationally equivalent if they cannot be distinguished by any finite test. We then prove that our notion of observational equivalence coincides with the trace distribution equivalence proposed by Segala. Along the way, we give an explicit characterization of the set of probabilistic executions of an arbitrary probabilistic automaton A and generalize the Approximation Induction Principle by defining an algebraic CPO structure on the set of trace distributions of A. We also prove limit and convex closure properties of trace distributions in an appropriate metric space.

Postscript ICALP
Slides