知识卡片

普通读书笔记卡 · 9202.c · 生成模型:GPT-6

归一化常数由全部候选加权得到

理解等级:中级

离散观测边际P(z)=Σx P(z|x)P(x),连续状态则为积分。原文内嵌式漏先验且末尾悬空乘号,不能只把各位置似然相加。只写比例关系可暂省分母,但输出概率时仍需正确归一。

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

      Let’s start with the third requirement, which might be the most confusing part of information we need to collect. It may make more sense to consider , the probability that we’d see this feature in a given location for every possible location. It is also common to see this term set to 1, with P(loc| feat) written as being proportional to the numerator of equation

    参考来源

    - 位置:《Introduction to Autonomous Robots Mechanisms, Sensors, Actuators, and Algorithms (Nikolaus Correll, Bradley Hayes etc.)》第16章《定位》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Introduction to Autonomous Robots Mechanisms, Sensors, Actuators, and Algorithms (Nikolaus Correll, Bradley Hayes etc.)/_epub-src/OEBPS/xhtml/chapter_16.xhtml) - 结论依据:摘录中关于归一化常数由全部候选加权得到的定义、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:Let’s start with the third requirement, which might be the most confusing part of information we need to collect. It may make more sense to consider , the probability that we’d see this feature in a given location for every possible location. It is also common to see this term set to 1, with P(loc| feat) written as being proportional to the numerator of equation