知识卡片
过高次数的拟合会追随噪声
当设计矩阵可逆(多项式基通常需样本横坐标互异)时,用与数据点一样多的系数可精确插值,却也可能吸收测量噪声,使未见样本的预测变差。
若矩阵秩不足,n=m也不保证能穿过所有样本;可先选较少基函数再最小二乘拟合。
这张卡片看不懂?
这张卡片的反馈已满 10 条,暂时无法继续提交。
已有反馈 · 0 / 10 条
还没有人提交反馈。
在整书笔记中的位置
原书出处 · 1 处
- 第28章 矩阵运算 · 查看原书依据
Such a high-degree polynomial F “fits the noise” as well as the data, however, and generally gives poor results when used to predict y for previously unseen values of x. It is usually better to choose n significantly smaller than m and hope
参考来源
- 位置:《算法导论(第4版)》第28章《矩阵运算》(OEBPS/chapter028.xhtml,文本字符[35000,40000))。
- 结论依据:第28章最小二乘段落以 n=m 说明高次多项式会拟合噪声,随后定义设计矩阵 A 与误差 Ac−y;精确插值还要求方程组相容,n=m 且 A 可逆时成立。
- 原始内容:Such a high-degree polynomial F “fits the noise” as well as the data, however, and generally gives poor results when used to predict y for previously unseen values of x . It is usually better to choose n significantly smaller than m and hope