概率图模型课本笔记（二）

最新推荐文章于 2025-09-07 19:29:44 发布

转载最新推荐文章于 2025-09-07 19:29:44 发布 · 75 阅读

0 ·

CC 4.0 BY-SA版权

原文链接：http://www.cnblogs.com/Wanggcong/p/6646186.html

文章标签：

#人工智能

无向图模型：

We have now presented three sets of independence assertions associated with a network structure H. For general distributions, I_p(H) is strictly weaker than I_l(H), which in turn is strictly weaker than I(H). However, all three definitions are equivalent for positive distributions.

I_p(H): pairwise assumption

I_l(H): local assumption

I(H): global assumption

转载于:https://www.cnblogs.com/Wanggcong/p/6646186.html