知识卡片
概率推断改变查询分布
模型可能容易从P(Z,Y)联合采样,而实际问题需要给定Y后的P(Z|Y)。概率推断在模型假设下计算或近似目标分布,不必等同于学习模型参数;生成、参数学习和给定证据的推断需要分开描述。
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
- · 概率生成建模入门
原书出处 · 1 处
- · 查看原书依据
2.2.5 Programming probabilistic inference Suppose we implement in code a random process that generates an outcome {x, y, z} from P (X, Y, Z ) as follows: Further, suppose we are interested in generating from P (Z |Y = 3). How might we do this? Our process can sample from P (Z ), P (X |Z ), and P (Y |Z ), but it is
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:摘录中关于概率推断改变查询分布的定义、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:2.2.5 Programming probabilistic inference Suppose we implement in code a random process that generates an outcome {x, y, z} from P (X, Y, Z ) as follows: Further, suppose we are interested in generating from P (Z |Y = 3). How might we do this? Our process can sample from P (Z ), P (X |Z ), and P (Y |Z ), but it is