知识卡片
相同图结构不抹除额外证据
反事实查询增加事实结果或另一潜在结果条件后,即使所需图的骨架相同,目标分布也可能改变。必须保留全部条件并按查询识别,不能因为图看起来一样就删除证据;给定结果也不意味着所有此类查询必然不可识别。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
- · 识别与因果层级
原书出处 · 1 处
- · 查看原书依据
conditions={T: +t} #2 ) #1 Hypothetical outcome A T=–t = +a #2 Factual condition T = +t This will produce a rather verbose level 2 estimand. We can then apply level 2 graphical identification algorithms to get a level 1 estimand, which will simplify to the following: I’ll show a simple derivation in the next section. For now, the intuition is that we are applying
参考来源
- 位置:《Causal AI》第10章《识别与因果层级》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Causal AI/_epub-src/OEBPS/Text/chapter-10.xhtml)
- 结论依据:摘录中关于相同图结构不抹除额外证据的定义、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:conditions={T: +t} #2 ) #1 Hypothetical outcome A T=–t = +a #2 Factual condition T = +t This will produce a rather verbose level 2 estimand. We can then apply level 2 graphical identification algorithms to get a level 1 estimand, which will simplify to the following: I’ll show a simple derivation in the next section. For now, the intuition is that we are applying