Bisimulation Metrics for Continuous Markov Decision Processes
From MaRDI portal
Publication:3225169
DOI10.1137/10080484XzbMath1253.39018MaRDI QIDQ3225169
Prakash Panangaden, Doina Precup, Norm Ferns
Publication date: 15 March 2012
Published in: SIAM Journal on Computing (Search for Journal in Brave)
linear programming; Markov decision process; reinforcement learning; bisimulation; metrics; continuous; statistical sampling
60J25: Continuous-time Markov processes on general state spaces
91G80: Financial applications of other theories
37H10: Generation, random and stochastic difference and differential equations
39A30: Stability theory for difference equations
39A50: Stochastic difference equations
Related Items
Probabilistic Model Checking of Labelled Markov Processes via Finite Approximate Bisimulations, Bisimulation for Markov Decision Processes through Families of Functional Expressions, Random Measurable Selections, A pseudometric in supervisory control of probabilistic discrete event systems, Adaptive aggregation for reinforcement learning in average reward Markov decision processes, Polynomial-time algorithms for computing distances of fuzzy transition systems, Weak bisimulation is sound and complete for pCTL\(^*\), An algebraic approach for inferring and using symmetries in rule-based models, Pseudometrics for State Aggregation in Average Reward Markov Decision Processes