知识卡片
溯因先更新外生联合分布
反事实推断先依据事实证据e,求外生量联合后验P(U|e),再在假想机制下传播。目标是联合后验,不能默认把每个噪声分别更新后独立拼起来;事实中的共同结果可能让原本独立的外生量产生后验依赖。
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
原书出处 · 1 处
- · 查看原书依据
Abduction —Infer the distribution of the exogenous variables given the factual conditions. Action —Implement the hypothetical condition as an ideal intervention (graph surgery) in the hypothetical world. Prediction —Use the conditional distribution on the exogenous variables from step 1 to derive the distributions of the hypothetical outcomes. I’ll illustrate how we can perform these steps using the parallel world graph for our online gaming example,
参考来源
- 位置:《Causal AI》第09章《通用反事实推断算法》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Causal AI/_epub-src/OEBPS/Text/chapter-9.xhtml)
- 结论依据:摘录中关于溯因先更新外生联合分布的定义、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:Abduction —Infer the distribution of the exogenous variables given the factual conditions. Action —Implement the hypothetical condition as an ideal intervention (graph surgery) in the hypothetical world. Prediction —Use the conditional distribution on the exogenous variables from step 1 to derive the distributions of the hypothetical outcomes. I’ll illustrate how we can perform these steps using the parallel world graph for our online gaming example,