← 返回起始笔记位置

知识卡片

普通读书笔记卡 · 8964 · 生成模型:GPT-6

边缘化汇总其余状态

理解等级:中级

求节点xi的后验边缘分布,需要对其他所有隐藏变量的状态求和,连续变量则积分。它保留xi的不确定性并汇总多种联合解释,不能用“其余变量取最优值”替代。后者属于最大化,会给出不同运算和结果。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      For the case of discrete variables, we compute the marginal probability at a node by summing over the states at all the other nodes, (29.9) where the notation j\i means “all possible values of j except for j = i.” 29.5 Simple Example of Inference in a Graphical Model To gain intuition, let's calculate the marginal probability for a simple example. Consider the three-node Markov

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第29章《概率图模型》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)/_epub-src/OPS/xhtml/c029.xhtml) - 结论依据:摘录中关于边缘化汇总其余状态的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:For the case of discrete variables, we compute the marginal probability at a node by summing over the states at all the other nodes, (29.9) where the notation j\i means “all possible values of j except for j = i.” 29.5 Simple Example of Inference in a Graphical Model To gain intuition, let's calculate the marginal probability for a simple example. Consider the three-node Markov