基于半Markov决策过程的概率布尔网络模型

基金项目:

国家自然科学基金;高等学校博士学科点专项科研基金;广东省高校优秀青年创新人才培养计划资助项目

详细信息

A probabilistic Boolean network based on semi-Markov decision processes

摘要: 借助半Markov决策过程理论研究了概率布尔网络中的最优控制问题．将概率布尔网络的控制模型转化为首达目标的半Markov决策过程模型，通过选取最优控制输入使得和系统相关的目标函数最小，从而解决了一个最优控制问题，并举例说明所用方法的效用．
- 风险概率
Abstract: An optimal control problem in a PBN with the help of the theory of Semi-Markov decision processes is studied. Specifically, a control model for a PBN as a first passage model for semi-Markov decision processes is firstly formulated and then an optimal control problem is solved for choosing optimal control inputs in a PBN such that an objective function related to system behavior is minimal. In order to illustrate the validity of our approach, an example is also displayed.
- risk probability