-
Perpendicular Distance Between a Hyperplane and a PointMathematics 2021. 1. 13. 17:39
Hyperplane
A hyperplane in two dimensions: a line
e.g. $\beta_0 + \beta_1 X_1 + \beta_2 X_2 = 0$A hyperplane in two dimensions: a plane
e.g. $\beta_0 + \beta_1 X_1 + \beta_2 X_2 + \beta_3 X_3 = 0$Perpendicular Distance Between a Hyperplnae and a Point in Two Dimensions
Let a point be $(x_1, x_2)$, and the hyperplane be $\beta_0 + \beta_1 X_1 + \beta_2 X_2 = 0$ (line). Then the shortest distance from the point to the line is the perpendicular distance $d$.
Then, the perpendicular distance is defined as:
$$ d = \frac{| \beta_0 + \beta_1 x_1 + \beta_2 x_2 |}{\sqrt{ \beta_1^2 + \beta_2^2 }} $$
Perpendicular Distance Between a Hyperplnae and a Point in $p$ Dimensions
The above equation can easily be expanded for the $p$ dimensions. A point is defined as $(x_1, x_2, \cdots, x_p)$ and a hyperplane is defined as $\beta_0 + \beta_1 X_1 + \cdots + \beta_p X_p$. Then, the perpendicular distance $d$ is:
$$ d = \frac{| \beta_0 + \beta_1 x_1 + \cdots + \beta_p x_p |}{\sqrt{ \beta_1^2 + \cdots + \beta_p^2 }} $$
'Mathematics' 카테고리의 다른 글
Conditional Expectation (0) 2021.01.27 Eigendecomposition (spectral decomposition), Singular Value Decomposition (SVD) (0) 2021.01.07 Multivariate Gaussian Distribution (0) 2021.01.07