知识卡片
单世界图拆分自然与固定值
SWIG将被干预节点拆成保留原父节点的自然随机部分与把固定值传给子节点的部分,并给受影响后代加干预下标。它表达单世界潜在结果之间的独立关系;构图不要求在被干预单位身上实际知道其未干预自然取值。
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
Next, we’ll look at another graph-based approach called single-world intervention graphs. 10.7.3 Counterfactual identification with single-world intervention graphs Single-world intervention graphs (SWIGs) provide an alternative to counterfactual identification with counterfactual graphs. Like a counterfactual graph, we construct a SWIG using the original causal DAG and the causal query. We’ll use the Netflix recommendation example to construct a SWIG for the interventions do(T = –t )
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:摘录中关于单世界图拆分自然与固定值的定义、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:Next, we’ll look at another graph-based approach called single-world intervention graphs. 10.7.3 Counterfactual identification with single-world intervention graphs Single-world intervention graphs (SWIGs) provide an alternative to counterfactual identification with counterfactual graphs. Like a counterfactual graph, we construct a SWIG using the original causal DAG and the causal query. We’ll use the Netflix recommendation example to construct a SWIG for the interventions do(T = –t )