知识卡片

普通读书笔记卡 · 9235 · 生成模型:GPT-6

有放回重抽近似抽样分布

理解等级:中级

从经验样本有放回重抽,常取与原样本同大小,重复计算统计量以研究其变化。不是必须抽更小样本,也不保证总比原估计更好;依赖数据需要相应重抽单位与方法。

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

      The method involves iterative resampling of a data set with replacement. Resampling instead of only estimating the statistics once on complete data set, provides better results. Repeating this multiple times helps to obtain a vector of estimates. Bootstrapping calculates the variance, expected value, and other relevant statistics of these estimates. Signal and Noise? In ML, signal is the true underlying pattern that we wish to

    参考来源

    - 位置:《Machine Learning Fundamentals: Concepts, Models, and Applications》第02章《机器学习基础》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Machine Learning Fundamentals: Concepts, Models, and Applications/_epub-src/OEBPS/xhtml/13_chapter02.xhtml) - 结论依据:摘录中关于有放回重抽近似抽样分布的定义、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:The method involves iterative resampling of a data set with replacement. Resampling instead of only estimating the statistics once on complete data set, provides better results. Repeating this multiple times helps to obtain a vector of estimates. Bootstrapping calculates the variance, expected value, and other relevant statistics of these estimates. Signal and Noise? In ML, signal is the true underlying pattern that we wish to