Approximating Labelled Markov Processes Again!
Labelled Markov processes are continuous-state fully probabilistic labelled transition systems. They can be seen as co-algebras of a suitable monad on the category of measurable space. The theory as developed so far included a treatment of bisimulation, logical characterization of bisimulation, weak bisimulation, metrics, universal domains for LMPs and approximations. Much of the theory involved delicate properties of analytic spaces.
Recently a new kind of averaging procedure was used to construct approximations. Remarkably, this version of the theory uses a dual view of LMPs and greatly simplifies the theory eliminating the need to consider aanlytic spaces. In this talk I will survey some of the ideas that led to this work.
KeywordsMarkov Process Conditional Expectation Markov Decision Process Ultrametric Space Markov Kernel
Unable to display preview. Download preview PDF.
- 2.Blute, R., Desharnais, J., Edalat, A., Panangaden, P.: Bisimulation for labelled Markov processes. In: Proceedings of the Twelfth IEEE Symposium On Logic In Computer Science, Warsaw, Poland (1997)Google Scholar
- 7.de Vink, E., Rutten, J.J.M.M.: Bisimulation for probabilistic transition systems: A coalgebraic approach. In: Proceedings of the 24th International Colloquium On Automata Languages And Programming (1997)Google Scholar
- 11.Desharnais, J., Gupta, V., Jagadeesan, R., Panangaden, P.: The metric analogue of weak bisimulation for labelled Markov processes. In: Proceedings of the Seventeenth Annual IEEE Symposium On Logic In Computer Science, pp. 413–422 (2002)Google Scholar
- 12.Kemeny, J.G., Snell, J.L.: Finite Markov Chains. Van Nostrand (1960)Google Scholar