知识卡片

普通读书笔记卡 · 9216.f · 生成模型:GPT-6

特征值刻画主轴方向的方差

理解等级:初级

协方差矩阵满足Cu=λu,λ对应单位主轴方向的方差,标准差为√λ。图中箭头按√λ缩放展示分布方向;估计协方差用1/N还是1/(N−1)需说明约定,不能把特征值直接当标准差。

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

      One can show that the principal components are eigenvectors of the data’s covariance matrix. For this, we need to compute the mean and variance of data as shown in figure B.1 across each dimension, shift the data so that it has zero mean, and then calculate the data’s covariance matrix. One can also show that the values of the corresponding eigenvalues are proportional to the

    参考来源

    - 位置:《Introduction to Autonomous Robots Mechanisms, Sensors, Actuators, and Algorithms (Nikolaus Correll, Bradley Hayes etc.)》附录B《附录B:线性代数》(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/appendixB.xhtml) - 结论依据:摘录中关于特征值刻画主轴方向的方差的定义、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:One can show that the principal components are eigenvectors of the data’s covariance matrix. For this, we need to compute the mean and variance of data as shown in figure B.1 across each dimension, shift the data so that it has zero mean, and then calculate the data’s covariance matrix. One can also show that the values of the corresponding eigenvalues are proportional to the