知识卡片

普通读书笔记卡 · 6538.c · 生成模型:GPT-6

变量消除控制中间因子规模

理解等级:中级

变量消除利用因子分解,对查询不需要的变量逐步求和或积分,再组合剩余因子。消除顺序影响中间因子大小及计算成本;使用图结构不代表算法会自动找到全局最优顺序。

这张卡片看不懂?

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

      Variable elimination uses the graph structure to optimally sum over the variables you wish to eliminate until the resulting table represents the target distribution. In contrast, belief propagation is a message-passing system; the graph is used to form different “cliques” of neighboring variables. For example, if P (Z |Y = 1) is the target distribution, Y=1 is a message iteratively passed back and forth between

    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:摘录中关于变量消除控制中间因子规模的定义、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:Variable elimination uses the graph structure to optimally sum over the variables you wish to eliminate until the resulting table represents the target distribution. In contrast, belief propagation is a message-passing system; the graph is used to form different “cliques” of neighboring variables. For example, if P (Z |Y = 1) is the target distribution, Y=1 is a message iteratively passed back and forth between