What is decidable about partially observable Markov decision processes with ω-regular objectives

• Published in: Journal of computer and system sciences Vol. 82; no. 5; pp. 878 - 911
• Main Authors: Chatterjee, Krishnendu, Chmelík, Martin, Tracol, Mathieu
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.08.2016
• Subjects: Complexity; Finite-memory strategies; Languages; Markov decision processes; Markov processes; Optimization; Parity; Parity objectives; Partially observable Markov decision processes (POMDPs); Qualitative analysis; Specifications; Strategy; ω-Regular conditions; Parity objectives; Finite-memory strategies; ω-Regular conditions; Markov decision processes; Partially observable Markov decision processes (POMDPs)

•Decidability of qualitative analysis of parity POMDPs under finite-memory strategies.•Optimal memory bounds and complexity (EXPTIME-completeness) for the above problem.•Implementation of our algorithm with several heuristics and experimental results. We consider partially observable Markov decision...