Bounds and good policies in stationary finite–stage Markovian decision problems
From MaRDI portal
Publication:3879083
DOI10.2307/1426499zbMath0437.90098MaRDI QIDQ3879083
Publication date: 1980
Published in: Advances in Applied Probability (Search for Journal in Brave)
Full work available at URL: https://doi.org/10.2307/1426499
infinite horizon; finite horizon; Borel state space; Borel action space; different decision horizons; first-step improvement method; interpolation between horizons; Markovian stationary decision problem; measurability and boundedness assumptions; sequential similarity transformation method
90C47: Minimax problems in mathematical programming
90C40: Markov and semi-Markov decision processes
Related Items
A unified approach to adaptive control of average reward Markov decision processes, Bounds for the quality and the number of steps in Bellman's value iteration algorithm