← 返回起始笔记位置

知识卡片

普通读书笔记卡 · 8977.b · 生成模型:GPT-6

亲和度与距离方向相反

理解等级:中级

亲和度通常越大表示越应该同组,距离通常越小表示越相似。使用颜色距离构建亲和度时,需明确是负距离、核函数还是阈值映射,不能同时把“大距离”和“高亲和”解释为同组。亲和函数也不必满足严格度量的所有公理。

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

      Other clustering methods are based on the idea of affinity, where we try to model a distance metric (the affinity) that tells us whether two items should be grouped. The distance metric usually measures how similar two patterns are, or how likely they are to be part of the same object.

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第31章《知觉分组》(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/c031.xhtml) - 结论依据:摘录中关于亲和度与距离方向相反的定义、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:Other clustering methods are based on the idea of affinity, where we try to model a distance metric (the affinity) that tells us whether two items should be grouped. The distance metric usually measures how similar two patterns are, or how likely they are to be part of the same object.