Simulation optimization algorithm for CTMDPs based on randomized stationary policies

Zidonghua Xuebao/Acta Automatica Sinica

Volumn 30, Issue 2, 2004, Pages 229-234

  Tang, Hao ^a,b   Xi, Hong Sheng ^a   Yin, Bao Qun ^a  

^a UNIVERSITY OF SCIENCE AND TECHNOLOGY OF CHINA   (China)

^b HEFEI UNIVERSITY OF TECHNOLOGY   (China)

Author keywords

Neuro dynamic programming; Performance potentials; Simulation optimization

Indexed keywords

ALGORITHMS; COMPUTER SIMULATION; ESTIMATION; OPTIMIZATION; THEOREM PROVING;

CONTINUOUS TIME MARKOV DECISION PROCESSES; LARGE SCALE STATE SPACE; NEURO DYNAMIC PROGRAMMING; PERFORMANCE POTENTIALS; RANDOMIZED STATIONARY POLICIES; SIMULATION OPTIMIZATION; SINGLE SAMPLE PATHS; UNIFORM MARKOV CHAINS;

MARKOV PROCESSES;

EID: 2942718962     PISSN: 02544156     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (7)

1. Cao X R, Chen H F. Perturbation realization, potentials and sensitivity analysis of Markov processes. IEEE Transactions on Automatic Control, 1997, 42(10): 1382-1393
2. Cao X R. Single sample path-based optimization of Markov chains. Journal of Optimization Theory and Applications, 1999, 100(3): 527-548
3. Bertsekas D P, Tsitsiklis J N. Neuro-Dynamic Programming. Belmont, MA: Athena Scientific, 1996
4. Marbach P, Tsitsiklis J N. Simulation-based optimization of Markov reward process. IEEE Transactions on Automatic Control, 2001, 46(2): 191-209
5. Tang H, Xi H S, Yin B Q. An on-line optimization algorithm for Markov control processes based on a single sample path. Control Theory and Applications, 2002, 19(6): 865-871 (in Chinese)
6. Xi H S, Tang H, Yin B Q. Optimal policies for a continuous time MCP with compact action set. Acta Automatica Sinica, 2003, 29(2): 206-211
7. Liu Z K, Tu F S. Single sample path-based sensitivity analysis of Markov processes. IEEE Transactions on Automatic Control, 1999, 44(4): 872-875