• Citer

Multiple-environment Markov decision processes

par Raskin, Jean-François ;Sankur, Ocan
Référence Leibniz international proceedings in informatics, 29, page (531-543)
Publication Publié, 2014-12

Article révisé par les pairs

• Détails
• Contenu

Résumé :

We introduce Multi-Environment Markov Decision Processes (MEMDPs) which are MDPs with a set of probabilistic transition functions. The goal in an MEMDP is to synthesize a single controller strategy with guaranteed performances against all environments even though the environment is unknown a priori. While MEMDPs can be seen as a special class of partially observable MDPs, we show that several verification problems that are undecidable for partially observable MDPs, are decidable for MEMDPs and sometimes have even efficient solutions.