← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 9172 · 生成模型:GPT-6

熵描述分布不确定性

理解等级:中级

离散熵H(p)=−Σp_i log p_i,以0log0=0的极限约定处理零概率。固定类别数时均匀分布熵最大、单类确定分布为零;对数底数决定单位,可用自然对数或其他固定底,原书10.16多出的N不应照抄。

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

      Here, the logarithm can be of basis ten or two. In any case, the entropy function has a couple of interesting properties: First, the logarithm from 0 (negative infinity) to 1 is negative (this is why probabilities yield positive values). Second, the logarithm of 1 is zero—that is, a distribution with only one element (pc = 1) has the lowest possible entropy. Third, the lower

    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:摘录中关于熵描述分布不确定性的定义、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:Here, the logarithm can be of basis ten or two. In any case, the entropy function has a couple of interesting properties: First, the logarithm from 0 (negative infinity) to 1 is negative (this is why probabilities yield positive values). Second, the logarithm of 1 is zero—that is, a distribution with only one element (pc = 1) has the lowest possible entropy. Third, the lower