【24h】

ENTROPIC MEASUREMENTS OF COMPLEXITY FOR FULLY OBSERVABLE MARKOV DECISION PROCESSES

机译:完全可观察的马尔可夫决策过程的复杂度的熵测

获取原文
获取原文并翻译 | 示例

摘要

Complex dynamic systems are an active topic, and how to measure the complexity of dynamic decisions still remains to be an unresolved problem. In this paper, we apply Shannon's entropy as well as other entropy indices to measure the complexity for fully observable Markov decision processes, and extend the measurements of the complexity, uncertainty and unpredictability for Markov Chains to the measurements for Markov decision processes. We develop a methodology of information-theoretic complexity measurements for fully observable Markov decision processes.
机译:复杂的动态系统是一个活跃的话题,如何衡量动态决策的复杂性仍然是一个尚未解决的问题。在本文中,我们应用香农熵和其他熵指标来测量完全可观测的马尔可夫决策过程的复杂度,并将对马尔可夫链的复杂性,不确定性和不可预测性的度量扩展到对马尔可夫决策过程的度量。我们为完全可观的马尔可夫决策过程开发了一种信息理论复杂性度量的方法。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号