Yiksan0315's Blog

Multivariate Distribution

  • or
  • 다번량 분포

# Tag:

  • Source/KU_ML

Multivariate Distribution(다변량 분포)

여러 개의 확률 변수가 포함된 결합 분포 함수.

혹은 여러 변수가 포함된 데이터. 단일 관측치에 대해 두 개 이상의 변수를 동시에 측정한 데이터를 의미한다. 보통 하나의 instance가 차원의 vector로 표현된다.

Expectation

\mathbb{E}[X]=\sum\limits_\vec{x}\vec{x}P(\vec{x})=\begin{bmatrix} \sum_\vec{x}x_1P(\vec{x}) \\ \vdots \\ \sum_\vec{x}x_dP(\vec{x}) \end{bmatrix}=\begin{bmatrix} \sum_{x_1}x_1P(x_1) \\ \vdots \\ \sum_{x_d}x_{d}P(x_d) \end{bmatrix}= \begin{bmatrix} \mu_{1} \\ \vdots \\ \mu_{d} \end{bmatrix}=\vec{\mu}

:평균을 계산할 때, 와 관련한 항만 상관이 있으므로 로 치환한다.

sample(train Data )에 대해서는, 으로 표기한다.

Covariance Matrix

\mathbb{CV}[X]\equiv\mathbb{E}[(X-\vec{\mu})(X-\vec{\mu})^\top]=\sum\limits_\vec{x}\begin{bmatrix} &\vdots \\ \cdots & (x_i-\mu_{i})(x_{j}- \mu_{j)}& \cdots \\ & \vdots \end{bmatrix}P(\vec{x})=\begin{bmatrix} \sigma_{11}&\cdots \sigma_{1d} \\ \vdots & \ddots & \vdots \\ \sigma_{d1} & \cdots & \sigma_{dd} \end{bmatrix}\equiv\mathbf{\Sigma}

이 때의 Covariance Matrix는, Symmetric-Positive-definite하게 정의되므로, Eigen decomposition이 항상 가능하고, 이 성립한다.

sample(train Data )에 대해서는, 으로 표기한다.
와 같이 계산된다.

Correlation Matrix

이 때, corrleation matrix는

: 즉, 로 표현된다

toc test

이 페이지는 리디주식회사에서 제공한 리디바탕 글꼴이 사용되어 있습니다. 리디바탕의 저작권은 리디주식회사가 소유하고 있습니다.

This Font Software is licensed under the SIL Open Font License, Version 1.1.

Copyright 2025. yiksan0315 All rights reserved.