Optimal decisions for continuous time Markov decision processes over finite planning horizons
详细信息 查看全文
- 作者：Peter Buchholz ; peter.buchholz@cs.tu-dortmund.de" class="auth_mail" title="E-mail the corresponding author ; Iryna Dohndorf iryna.dohndorf@cs.tu-dortmund.de" class="auth_mail" title="E-mail the corresponding author ; Dimitri Scheftelowitsch dimitri.scheftelowitsch@cs.tu-dortmund.de" class="auth_mail" title="E-mail the corresponding author
- 关键词：Continuous time Markov decision process ; Finite horizon ; Uniformization ; Numerical techniques ; Optimization
- 刊名：Computers and Operations Research
- 出版年：2017
- 出版时间：January 2017
- 年：2017
- 卷：77
- 期：Complete
- 页码：267-278
- 全文大小：1009 K
文摘
- •
A new algorithm to compute accumulated rewards for Continuous Time Markov Decision Processes with action dependent rewards over finite horizons.
- •
A proof that the algorithm guarantees a global error in O(δ) for time step δ.
- •
Experimental comparision of available algorithms to analyze accumulated rewards for Continuous Time Markov Decision Processes with action dependent rewards over finite horizons.