Welcome to Randal Douc's wiki

A collaborative site on maths but not only!

User Tools

Site Tools


world:kullback

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Next revision
Previous revision
world:kullback [2023/11/10 10:08]
rdouc created
world:kullback [2023/11/10 10:19] (current)
rdouc [Proof]
Line 1: Line 1:
 {{page>:​defs}} {{page>:​defs}}
  
-The aim is to compute the Kullback divergence ​between two normal distributions+====== ​Kullback-Leibler ​divergence ​for normal distributions ​====== 
 + 
 +<WRAP center round tip 90%> 
 +Let $\mu_0,​\mu_1 \in \rset^p$ and $ \Sigma_0,​\Sigma_1 ​ \in \rset^{p \times p}$ where $\Sigma_0,​\Sigma_1$ are symmetric definite positive. Then,  ​
 $$ $$
-\klbck{\N(\mu_0,​\Sigma_0)}{\N(\mu_1,​\Sigma_1)}+\klbck{\N(\mu_0,​\Sigma_0)}{\N(\mu_1,​\Sigma_1)}=-\frac{p}{2} ​ + \frac{(\mu_0-\mu_1)^T \Sigma_1^{-1} (\mu_0-\mu_1) }{2} + \frac{Tr\lr{\Sigma_1^{-1} \Sigma_0}}{2} - \frac{1}{2} \log \frac{\mathrm{det} \Sigma_0}{\mathrm{det} \Sigma_1}
 $$ $$
-where $\mu_0,​\mu_1 \in \rset^p$ and $ \Sigma_0,​\Sigma_1 ​ \in \rset^{p \times p}$. + 
 +</​WRAP>​ 
 + 
 +===== Proof =====
  
 Assume that $X_0\sim \N(\mu_0,​\Sigma_0)$ then $X_0= \mu_0 + \Sigma_0^{1/​2} U_0$ where $U_0\sim \N(0,I_p)$ Assume that $X_0\sim \N(\mu_0,​\Sigma_0)$ then $X_0= \mu_0 + \Sigma_0^{1/​2} U_0$ where $U_0\sim \N(0,I_p)$
Line 14: Line 20:
     &= -\frac{p}{2} ​ + \frac{(\mu_0-\mu_1)^T \Sigma_1^{-1} (\mu_0-\mu_1) }{2} + \frac 1 2 \PE\lrb{U_0^T \Sigma_0^{1/​2} \Sigma_1^{-1} \Sigma_0^{1/​2} U_0} - \frac{1}{2} \log \frac{\mathrm{det} \Sigma_0}{\mathrm{det} \Sigma_1}\\ ​     &= -\frac{p}{2} ​ + \frac{(\mu_0-\mu_1)^T \Sigma_1^{-1} (\mu_0-\mu_1) }{2} + \frac 1 2 \PE\lrb{U_0^T \Sigma_0^{1/​2} \Sigma_1^{-1} \Sigma_0^{1/​2} U_0} - \frac{1}{2} \log \frac{\mathrm{det} \Sigma_0}{\mathrm{det} \Sigma_1}\\ ​
     &= -\frac{p}{2} ​ + \frac{(\mu_0-\mu_1)^T \Sigma_1^{-1} (\mu_0-\mu_1) }{2} + \frac{Tr\lr{\Sigma_1^{-1} \Sigma_0}}{2} - \frac{1}{2} \log \frac{\mathrm{det} \Sigma_0}{\mathrm{det} \Sigma_1}     &= -\frac{p}{2} ​ + \frac{(\mu_0-\mu_1)^T \Sigma_1^{-1} (\mu_0-\mu_1) }{2} + \frac{Tr\lr{\Sigma_1^{-1} \Sigma_0}}{2} - \frac{1}{2} \log \frac{\mathrm{det} \Sigma_0}{\mathrm{det} \Sigma_1}
 +\end{align*}
 +where in the last line, we have used that 
 +
 +\begin{align*}
 +    \PE\lrb{U_0^T \Sigma_0^{1/​2} \Sigma_1^{-1} \Sigma_0^{1/​2} U_0}&​=\lr{\PE\lrb{Tr\lr{U_0^T \Sigma_0^{1/​2} \Sigma_1^{-1} \Sigma_0^{1/​2} U_0}}}=Tr\lr{\PE\lrb{\Sigma_1^{-1} \Sigma_0^{1/​2} U_0 U_0^T \Sigma_0^{1/​2} }} \\ 
 +    &​=Tr\lr{\Sigma_1^{-1} \Sigma_0^{1/​2} \PE\lrb{U_0 U_0^T} \Sigma_0^{1/​2} }=Tr\lr{\Sigma_1^{-1} \Sigma_0}
 \end{align*} \end{align*}
world/kullback.1699607281.txt.gz · Last modified: 2023/11/10 10:08 by rdouc