知识卡片
拒绝筛选有支持与效率限制
从联合模型反复采样,只保留满足离散观察条件的结果,可近似相应条件分布。稀有条件会造成极低接受率,有限样本甚至一个都不剩;连续变量精确等于某值通常是零概率事件,不能直接照搬这种筛选。
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
For example, a simple class of probabilistic inference algorithms is called accept/reject algorithms. Applying a simple accept/reject technique to generating from P (Z |Y = 3) works as follows: Repeatedly generate {x, y, z} using our generator for P (X, Y, Z ). Throw away any generated outcome where y is not equal to 3. The resulting set of outcomes for Z will have the
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:摘录中关于拒绝筛选有支持与效率限制的定义、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:For example, a simple class of probabilistic inference algorithms is called accept/reject algorithms. Applying a simple accept/reject technique to generating from P (Z |Y = 3) works as follows: Repeatedly generate {x, y, z} using our generator for P (X, Y, Z ). Throw away any generated outcome where y is not equal to 3. The resulting set of outcomes for Z will have the