← 返回起始笔记位置

知识卡片

普通读书笔记卡 · 8963.c · 生成模型:GPT-6

高斯均值和众数重合

理解等级:中级

非退化高斯后验的均值与密度众数相同,所以平方误差MMSE与MAP给出相同向量。一般多峰或偏斜后验没有这个性质。即使点估计重合,保留后验协方差仍能提供单一向量无法表达的不确定性。

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 a Gaussian probability distribution, the MAP estimate and the MMSE estimate are always the same, since the mean of the distribution is always its maximum value. For now, we'll assume we seek the MMSE estimate. By the properties of the multivariate mean, to find the mean at each variable, or node, in a network, we can first find the marginal probability at each node,

    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:摘录中关于高斯均值和众数重合的定义、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 a Gaussian probability distribution, the MAP estimate and the MMSE estimate are always the same, since the mean of the distribution is always its maximum value. For now, we'll assume we seek the MMSE estimate. By the properties of the multivariate mean, to find the mean at each variable, or node, in a network, we can first find the marginal probability at each node,