# 最大熵模型中的对数似然函数的解释

${L}_{\overline{p}}=\prod _{x}p\left(x{\right)}^{\overline{p}\left(x\right)}.$

$L\left({x}_{1},{x}_{2},...,{x}_{n};\theta \right)=\prod _{i=1}^{n}p\left({x}_{i};\theta \right).$

$L\left({x}_{1},{x}_{2},...,{x}_{n};\theta \right)=\prod _{i=1}^{k}p\left({v}_{i};\theta {\right)}^{C\left(X={v}_{i}\right)}.$

$L\left({x}_{1},{x}_{2},...,{x}_{n};\theta {\right)}^{\frac{1}{n}}=\prod _{i=1}^{k}p\left({v}_{i};\theta {\right)}^{\frac{C\left(X={v}_{i}\right)}{n}}.$

$L\left({x}_{1},{x}_{2},...,{x}_{n};\theta {\right)}^{\frac{1}{n}}=\prod _{x}p\left(x;\theta {\right)}^{\overline{p}\left(x\right)}.$

$L\left(x;\theta \right)=\prod _{x}p\left(x:\theta {\right)}^{\overline{p}\left(x\right)}.$

$\begin{array}{}\text{(5)}& {L}_{\overline{p}}& =& \mathrm{log}\prod _{x,y}p\left(x,y{\right)}^{\overline{p}\left(x,y\right)}\text{(6)}& & =& \sum _{x,y}\overline{p}\left(x,y\right)\mathrm{log}p\left(x,y\right)\text{(7)}& & =& \sum _{x,y}\overline{p}\left(x,y\right)\mathrm{log}\left[\overline{p}\left(x\right)p\left(y|x\right)\right]\text{(8)}& & =& \sum _{x,y}\overline{p}\left(x,y\right)\mathrm{log}p\left(y|x\right)+\sum _{x,y}\overline{p}\left(x,y\right)\mathrm{log}\overline{p}\left(x\right)\end{array}$

${L}_{\overline{p}}=\sum _{x,y}\overline{p}\left(x,y\right)\mathrm{log}p\left(y|x\right).$