一种基于柯氏复杂度的因果网络定向方法
作者:
作者单位:

国防科技大学信息通信学院

作者简介:

通讯作者:

中图分类号:

TP181

基金项目:

军委科技委理论科研项目(19JSLLKY015)


A Causal Network Orientation Method based on Kolmogorov Complexity
Author:
Affiliation:

College of Information and Communication, National University of Defense Technology

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    因果网络定向问题实质是一个“多对多”因果关系发现过程,传统的V-结构定向方法只能确定一组马尔可夫等价类而非最终的因果关系。为解决该问题,从柯氏复杂度的因果推断原理视角出发,利用贝叶斯链式法则推导出局部网络因果定向规则,并在此基础上提出高维全局网络因果定向方法。同时,将前者运用于改进基于局部条件独立信息搜索学习马尔科夫毯典型算法,后者运用于改进基于约束的因果网络结构学习典型算法,实验结果表明改进后算法在保证较高准确率的同时有效提升执行效率。

    Abstract:

    The nature of causal network orientation problem is a "many-to-many" causal discovery process.The traditional V-structure method can only determine a set of Markov equivalent classes rather than the final causal relationship.In order to solve this problem,this paper offer a new method based on the Kolmogorov Complexity,including deducing a causal orientation rule of local network from the Bayesian chain rule,thus futher,proposing a high-dimensional global network causal orientation rule on this basis.At the same time,the former rule is used to improve Markov blanket typical algorithm based on the local condition independent information searching;the latter is used to improve the constraint based causal network structure learning typical algorithm.The experimental results show that the improved algorithm can effectively improve the execution efficiency while ensuring high accuracy.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2020-01-02
  • 最后修改日期:2020-06-06
  • 录用日期:2020-06-12
  • 在线发布日期:
  • 出版日期: