The paper establishes NP-hardness for optimal policy search in static DRMDPs under Markovian non-randomized and randomized policies, using ambiguity sets with two kernels and showing sub-optimal strict local minimizers for the latter.
Robust markov decision processes.Mathematics of Operations Research, 38(1):153–183, 2013
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.OC 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Computational Hardness of Static Distributionally Robust Markov Decision Processes
The paper establishes NP-hardness for optimal policy search in static DRMDPs under Markovian non-randomized and randomized policies, using ambiguity sets with two kernels and showing sub-optimal strict local minimizers for the latter.