← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 6566.b · 生成模型:GPT-6

固定常数不等于观察到的独立

理解等级:中级

do(X=x)后X为常数,即使图中仍有X到Y的有向路径,也不能据d连接宣称固定X与Y在该分布里统计相关。因果效应通过比较不同x下的结果分布表达;d连接只表示图不保证独立,不能反推每条路径都有非零相关。

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

      Only outgoing paths to other nodes remain. As a result, the remaining paths from that variable reflect dependence due to that variable’s causal influence on other variables. The ideal intervention removes the causal influence the target variable receives from its direct parents. Thus, it removes any dependence on other variables that flows through those parents.

    参考来源

    - 位置:《Causal AI》第07章《干预与因果效应》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Causal AI/_epub-src/OEBPS/Text/chapter-7.xhtml) - 结论依据:摘录中关于固定常数不等于观察到的独立的定义、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:Only outgoing paths to other nodes remain. As a result, the remaining paths from that variable reflect dependence due to that variable’s causal influence on other variables. The ideal intervention removes the causal influence the target variable receives from its direct parents. Thus, it removes any dependence on other variables that flows through those parents.