Modeling Markov Decision Processes with Imprecise Probabilities Using Probabilistic Logic Programming
[edit]
Proceedings of the Tenth International Symposium on Imprecise Probability: Theories and Applications, PMLR 62:4960, 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 Setvalued Transitions (MDPSTs) we then obtain the complexity of onestep inference for the resulting MDPIPs and MDPSTs. We also present results of independent interest on the complexity of inference with probabilistic logic programs containing intervalvalued probabilistic assessments. Finally, we also discuss policy generation techniques.
Related Material


