Markov Decision Processes from Colored Petri Nets
Models that are suitable for planning are not always easy to specify. In this paper we investigate the conversion of Petri nets into factored Markov decision processes: the former are relatively easy to build while the latter are adequate for policy generation. To represent probabilities that are needed when planning under uncertainty, we introduce factored Petri nets; we then describe the conversion of factored Petri nets in Markov decision processes.
KeywordsPlanning under uncertainty Markov decision processes Colored Petri nets
Unable to display preview. Download preview PDF.
- 3.Hass, P.J.: Stochastic Petri nets for modeling and simulation. In: Winter Simulation Conference, Washington, pp. 101–112 (2004)Google Scholar
- 5.Jensen, K.: An introduction to the practical use of colored Petri nets, pp. 237–292. Springer, Berlin (1998)Google Scholar
- 6.Jensen, K., Kristensen, L.M., Cristensen, S.: The Practioners Guide to Coloured Petri Nets, pp. 98–132. Springer, Berlin (1998)Google Scholar
- 7.Miyagi, P.E.: Controle Programavel. Editora Edgard Blcher (1996)Google Scholar
- 8.Russel, S., Norvig, P.: Inteligencia Artificial. Editora Campus (2004)Google Scholar
- 10.Trevisan, F.W., Cozman, F.G., Barros, L.N.: Planning under risk and Knightian uncertainty. In: International Joint Conference on Artificial Intelligence, pp. 2023–2028 (2007)Google Scholar
- 11.Delgado, K.V., Barros, L.N., Cozman, F.G., Shirota, R.: Representing and solving factored Markov decision processes with imprecise probabilities. In: 6th International Symposium on Imprecise Probability: Theories and Applications, Durham, United Kingdom, pp. 169–178 (2009)Google Scholar