Information Decomposition on Structured Space
2016 IEEE International Symposium on Information Theory (ISIT)（2016）
摘要
We build information geometry for a partially ordered set of variables and define the orthogonal decomposition of information theoretic quantities. The natural connection between information geometry and order theory leads to efficient decomposition algorithms. This generalization of Amari's seminal work on hierarchical decomposition of probability distributions on event combinations enables us to analyze high-order statistical interactions arising in neuroscience, biology, and machine learning.
更多查看译文
关键词
information decomposition,structured space,information geometry,partially ordered set,orthogonal decomposition,information theoretic quantities,order theory,decomposition algorithms,hierarchical decomposition,probability distributions,event combinations,high-order statistical interactions
AI 理解论文
溯源树
样例
生成溯源树，研究论文发展脉络
Chat Paper
正在生成论文摘要