Operations Research · 2005 · 140 citations · 16 references
BusinessComputer ScienceDecision SciencesAlgorithmsInventory ControlMarkov Decision ProcessesAdaptive ControlOptimal ControlReinforcement LearningReinforcement Learning FoundationsStochastic ControlAdaptive SamplingAdaptive Sampling StrategiesFinite Horizon ControlFinite-horizon ApproximationFinite-horizon Decision MakingFinite-horizon Decision ProcessesReceding Horizon ControlWorst-case Complexity
Based on recent results for multiarmed bandit problems, we propose an adaptive sampling algorithm that approximates the optimal value of a finite-horizon Markov decision process (MDP) with finite state and action spaces. The algorithm adaptively chooses which action to sample as the sampling process proceeds and generates an asymptotically unbiased estimator, whose bias is bounded by a quantity that converges to zero at rate (ln N)/N, where N is the total number of samples that are used per state sampled in each stage. The worst-case running-time complexity of the algorithm is O((|A|N) H ), independent of the size of the state space, where |A| is the size of the action space and H is the horizon length. The algorithm can be used to create an approximate receding horizon control to solve infinite-horizon MDPs. To illustrate the algorithm, computational results are reported on simple examples from inventory control.
16
Adhi Harmoko S, M.Komp, Joseph Marie Jacquard et al. · 2005 · 18.4K citations
V. J. Rayward‐Smith, Thomas H. Cormen, Charles E. Leiserson et al. · Journal of the Operational Research Society · 1991 · 17K citations