← 返回起始笔记位置

知识卡片

普通读书笔记卡 · 8962.a · 生成模型:GPT-6

固定观测将证据纳入局部势

理解等级:中级

观测y确定后,将相关势函数中的y代入,得到关于隐藏变量x的局部证据,再归一化形成p(x|y)。若势函数乘积原本未归一化,分母应按条件配分函数处理,不能未经定义就把它等同数值上的p(y)。

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 brevity, we write p(y) for p(y1, y2, y3). Thus, to form the conditional distribution, within a normalization factor, we simply include the observed variable values into the joint probability. For vision applications, we often use such Markov chain structures to describe events over time.

    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:摘录中关于固定观测将证据纳入局部势的定义、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 brevity, we write p(y) for p(y1, y2, y3). Thus, to form the conditional distribution, within a normalization factor, we simply include the observed variable values into the joint probability. For vision applications, we often use such Markov chain structures to describe events over time.