知识卡片
消息是收件节点的函数
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?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
- · 概率图模型
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,