← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 9172.b · 生成模型:GPT-6

KL是交叉熵减去目标熵

理解等级:中级

在适当支持条件下KL(p||q)=H(p,q)−H(p)≥0,p=q时可等于零。原文邻句把差的次序说反且称交叉熵永远更大需改为不小于;KL一般不对称,不能把它当普通距离度量。

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

      This difference between the entropy of the true distribution and the cross-entropy between the true and the estimated distribution is known as Kullback-Leibler divergence. It is a measure of dissimilarity between two distributions.

    参考来源

    - 位置:《Introduction to Autonomous Robots Mechanisms, Sensors, Actuators, and Algorithms (Nikolaus Correll, Bradley Hayes etc.)》第10章《人工神经网络》(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_10.xhtml) - 结论依据:摘录中关于KL是交叉熵减去目标熵的定义、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:This difference between the entropy of the true distribution and the cross-entropy between the true and the estimated distribution is known as Kullback-Leibler divergence. It is a measure of dissimilarity between two distributions.