知识卡片

普通读书笔记卡 · 6539.e · 生成模型:GPT-6

因果生成到观测是多对一

理解等级:中级

不同DGP可能产生相同完整联合分布,不同完整分布又可能在隐藏变量被边缘化后产生相同观测分布。有限数据进一步损失信息。因此增加同类观察数据不能自动区分所有因果机制;还需要结构假设或新的干预、测量信息。

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

      In summary, the DGP entails the full joint distribution, and marginalizing over the full joint distribution produces the observational joint distribution. Sampling from that distribution produces the observed data and the corresponding empirical joint distribution. There is a many-to-one relationship as we move down this hierarchy that has implications for causal modeling and inference.

    参考来源

    - 位置:《Causal AI》第02章《概率生成建模入门》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Causal AI/_epub-src/OEBPS/Text/chapter-2.xhtml) - 结论依据:摘录中关于因果生成到观测是多对一的定义、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:In summary, the DGP entails the full joint distribution, and marginalizing over the full joint distribution produces the observational joint distribution. Sampling from that distribution produces the observed data and the corresponding empirical joint distribution. There is a many-to-one relationship as we move down this hierarchy that has implications for causal modeling and inference.