Modeling Markov Decision Processes with Imprecise Probabilities Using Probabilistic Logic Programming

Thiago P. Bueno, Denis D. Mauá, Leliane N. Barros, Fabio G. Cozman
; Proceedings of the Tenth International Symposium on Imprecise Probability: Theories and Applications, PMLR 62:49-60, 2017.

Abstract

We study languages that specify Markov Decision Processes with Imprecise Probabilities (MDPIPs) by mixing probabilities and logic programming. We propose a novel language that can capture MDPIPs and Markov Decision Processes with Set-valued Transitions (MDPSTs) we then obtain the complexity of one-step inference for the resulting MDPIPs and MDPSTs. We also present results of independent interest on the complexity of inference with probabilistic logic programs containing interval-valued probabilistic assessments. Finally, we also discuss policy generation techniques.

Cite this Paper


BibTeX
@InProceedings{pmlr-v62-bueno17a, title = {Modeling {M}arkov Decision Processes with Imprecise Probabilities Using Probabilistic Logic Programming}, author = {Thiago P. Bueno and Denis D. Mauá and Leliane N. Barros and Fabio G. Cozman}, booktitle = {Proceedings of the Tenth International Symposium on Imprecise Probability: Theories and Applications}, pages = {49--60}, year = {2017}, editor = {Alessandro Antonucci and Giorgio Corani and Inés Couso and Sébastien Destercke}, volume = {62}, series = {Proceedings of Machine Learning Research}, month = {10--14 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v62/bueno17a/bueno17a.pdf}, url = {http://proceedings.mlr.press/v62/bueno17a.html}, abstract = {We study languages that specify Markov Decision Processes with Imprecise Probabilities (MDPIPs) by mixing probabilities and logic programming. We propose a novel language that can capture MDPIPs and Markov Decision Processes with Set-valued Transitions (MDPSTs) we then obtain the complexity of one-step inference for the resulting MDPIPs and MDPSTs. We also present results of independent interest on the complexity of inference with probabilistic logic programs containing interval-valued probabilistic assessments. Finally, we also discuss policy generation techniques.} }
Endnote
%0 Conference Paper %T Modeling Markov Decision Processes with Imprecise Probabilities Using Probabilistic Logic Programming %A Thiago P. Bueno %A Denis D. Mauá %A Leliane N. Barros %A Fabio G. Cozman %B Proceedings of the Tenth International Symposium on Imprecise Probability: Theories and Applications %C Proceedings of Machine Learning Research %D 2017 %E Alessandro Antonucci %E Giorgio Corani %E Inés Couso %E Sébastien Destercke %F pmlr-v62-bueno17a %I PMLR %J Proceedings of Machine Learning Research %P 49--60 %U http://proceedings.mlr.press %V 62 %W PMLR %X We study languages that specify Markov Decision Processes with Imprecise Probabilities (MDPIPs) by mixing probabilities and logic programming. We propose a novel language that can capture MDPIPs and Markov Decision Processes with Set-valued Transitions (MDPSTs) we then obtain the complexity of one-step inference for the resulting MDPIPs and MDPSTs. We also present results of independent interest on the complexity of inference with probabilistic logic programs containing interval-valued probabilistic assessments. Finally, we also discuss policy generation techniques.
APA
Bueno, T.P., Mauá, D.D., Barros, L.N. & Cozman, F.G.. (2017). Modeling Markov Decision Processes with Imprecise Probabilities Using Probabilistic Logic Programming. Proceedings of the Tenth International Symposium on Imprecise Probability: Theories and Applications, in PMLR 62:49-60

Related Material