← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 8964.b · 生成模型:GPT-6

消息是收件节点的函数

理解等级:中级

从j发往i的消息mⱼᵢ(xᵢ)以收件节点状态xᵢ为自变量,概括发送侧子树对这些状态的支持。它不是发送节点自身的边缘概率,也不必归一化为概率。明确自变量有助于避免矩阵转置和索引方向错误。

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

      The partial sums of equation (29.13) are named messages because they pass information from one node to another. We call the message from node 3 to node 2, . The other partial sum is the message from node 2 to node 1, . Note that the messages are always messages about the states of the node that the message is being sent to, that is,

    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:摘录中关于消息是收件节点的函数的定义、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:The partial sums of equation (29.13) are named messages because they pass information from one node to another. We call the message from node 3 to node 2, . The other partial sum is the message from node 2 to node 1, . Note that the messages are always messages about the states of the node that the message is being sent to, that is,