Title :
A Survey of Approximate Dynamic Programming
Author :
Wang, Lin ; Peng, Hui ; Zhu, Hua-Yong ; Shen, Lin-Cheng
Author_Institution :
Coll. of Mechatron. Eng. & Autom., Nat. Univ. of Defense Technol., Changsha, China
Abstract :
Multi-stage decision problems under uncertainty are abundant in process industries. Markov decision process (MDP) is a general mathematical formulation of such problems. Whereas stochastic programming and dynamic programming are the standard methods to solve MDPs, their unwieldy computational requirements limit their usefulness in real applications. Approximate dynamic programming (ADP) combines simulation and function approximation to alleviate the "curse-of-dimensionality" associated with the traditional dynamic programming approach. In this paper, the method of ADP, which abates the curse-of-dimensionality by solving the DP within a carefully chosen, small subset of the state space, was introduced; a survey of recent research directions within the field of ADP had been made.
Keywords :
Markov processes; decision making; dynamic programming; function approximation; learning (artificial intelligence); Markov decision process; approximate dynamic programming; curse-of-dimensionality; function approximation; mathematical formulation; multistage decision problems; process industries; reinforcement learning; state space; stochastic programming; Cybernetics; Dynamic programming; Educational institutions; Function approximation; Intelligent systems; Learning; Least squares approximation; Man machine systems; State-space methods; Stochastic processes; Approximate Dynamic Programming; Markov Decision Processes; Reinforcement Learning;
Conference_Titel :
Intelligent Human-Machine Systems and Cybernetics, 2009. IHMSC '09. International Conference on
Conference_Location :
Hangzhou, Zhejiang
Print_ISBN :
978-0-7695-3752-8
DOI :
10.1109/IHMSC.2009.222
