知识卡片

普通读书笔记卡 · 6572 · 生成模型:GPT-6

必要性概率限定已发生人群

理解等级:中级

PN=P(Y0=0|X=1,Y=1):在原因与结果实际都发生的对象中,问若去掉原因结果是否不发生。它是特定事实群体中的反事实概率,不是总体风险差,也不是“X是唯一原因”的证明。

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

      probability of necessity (PN) is the query P (YX= 0 = 0|X = 1, Y = 1). In plain language, the question underlying PN is “For cases where X happened, and Y happened, if X had not happened, would Y not have happened?” In other words, did X need to happen for Y to happen?

    参考来源

    - 位置:《Causal AI》第08章《反事实与平行世界》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Causal AI/_epub-src/OEBPS/Text/chapter-8.xhtml) - 结论依据:摘录中关于必要性概率限定已发生人群的定义、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:probability of necessity (PN) is the query P (YX= 0 = 0|X = 1, Y = 1). In plain language, the question underlying PN is “For cases where X happened, and Y happened, if X had not happened, would Y not have happened?” In other words, did X need to happen for Y to happen?