← 返回起始笔记位置

知识卡片

普通读书笔记卡 · 9257 · 生成模型:GPT-6

概率归一不改变最大得分位置

理解等级:中级

Softmax以pᵢ=exp(zᵢ)/Σⱼexp(zⱼ)归一同一向量,输出正且总和1,最大分量位置保持不变。它不是使大小比较成为可能的必要操作;分母须随j变化,原式图分母误用固定yᵢ。

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

      Softmax: the softmax function is a mathematical function commonly used in neural networks to convert a vector of numbers into a vector of probabilities. A vector of probabilities allows us to choose the element with the highest value as the decision made by a neural network, whereas it is much more difficult to compare numbers that may be very small, large, positive, or negative, as

    参考来源

    - 位置:《Machine Learning Fundamentals: Concepts, Models, and Applications》第08章《深度学习模型》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Machine Learning Fundamentals: Concepts, Models, and Applications/_epub-src/OEBPS/xhtml/21_chapter08.xhtml) - 结论依据:摘录中关于概率归一不改变最大得分位置的定义、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:Softmax: the softmax function is a mathematical function commonly used in neural networks to convert a vector of numbers into a vector of probabilities. A vector of probabilities allows us to choose the element with the highest value as the decision made by a neural network, whereas it is much more difficult to compare numbers that may be very small, large, positive, or negative, as