Publication:3347664
From MaRDI portal
zbMath0558.90100MaRDI QIDQ3347664
Publication date: 1984
bang-bang optimal policies; existence of an optimal randomized stationary policy; finite state and compact action space; long-run average reward
49M29: Numerical methods involving duality
49L20: Dynamic programming in optimal control and differential games
90C39: Dynamic programming
93E20: Optimal stochastic control
49J30: Existence of optimal solutions belonging to restricted classes (Lipschitz controls, bang-bang controls, etc.)
90C40: Markov and semi-Markov decision processes
