← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 9235.d · 生成模型:GPT-6

方差度量预测的平方偏离

理解等级:中级

固定输入后,在重复训练样本下看预测的变化,方差为E[(预测−平均预测)²]。原文均值差公式漏平方会恒为零,不能用于方差计算;偏差则比较平均预测与目标函数。

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

      High variance means that the small changes in training data lead to big changes in the result. Variance is introduced with high sensitivity to variations in training data. Bias is the type of error that occurs due to wrong assumptions about data. For example, assuming that the data are linear when it follows some other complex function. Bias and variance are inversely related. It is

    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:摘录中关于方差度量预测的平方偏离的定义、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:High variance means that the small changes in training data lead to big changes in the result. Variance is introduced with high sensitivity to variations in training data. Bias is the type of error that occurs due to wrong assumptions about data. For example, assuming that the data are linear when it follows some other complex function. Bias and variance are inversely related. It is - 生成时纠错核验:[scikit-learn官方定义与示例](https://scikit-learn.org/stable/auto_examples/ensemble/plot_bias_variance.html)(2026-11-06;仅核验机制,未执行示例)。