Information decomposition on structured space
M Sugiyama, H Nakahara… - 2016 IEEE International …, 2016 - ieeexplore.ieee.org
2016 IEEE International Symposium on Information Theory (ISIT), 2016•ieeexplore.ieee.org
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.
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.
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.
ieeexplore.ieee.org
以上显示的是最相近的搜索结果。 查看全部搜索结果