Reference

https://en.wikibooks.org/wiki/LaTeX/Mathematics

Examples

We can integrate markdown between $ $ or $$ $$

Single $ will be inline

Double $$ will be centered into a new line

LaTeX	Markdow
$\widehat{q}$	`\widehat{q}`
$\|x-x_M\|$	`\|x-x_M\|`
$f_M(x)$	`f_M(x)`
$\displaystyle\sum_{m}$	`\displaystyle\sum_{m}`
$\langle{x,w_m}\rangle$	`\langle{x,w_m}\rangle`
$\\| v_m \\|$	`\\| v_m \\|`
$\cos$	`\cos`
$\mathbb{Z}^d$	`\mathbb{Z}$^d`
$\infty$	`\infty`
$f \in L^2$	`f \in L^2`
$$	`\implies`
$\lim\limits_{M \to \infty}$	`\lim\limits_{M \to \infty}`
$P(A\|B) = \frac{P(B\|A)*P(A)}{P(B)}$	`P(A\|B) = \frac{P(B\|A)*P(A)}{P(B)}`
$\begin{align} \\ Q_t(a) &= \frac{\text{sum of rewards when } \mathit{a} \text{ taken prior to }\mathit{t}}{\text{number of times } \mathit{a} \text{ taken prior to }\mathit{t}} \\ & = \frac{\displaystyle\sum_{i=1}^{t-1} R_i.\mathcal{1}_{A_i=a}}{\displaystyle\sum_{i=1}^{t-1} \mathcal{1}_{A_i=a}} \end{align}$	`\begin{align} \\` `Q_t(a) &= \frac{\text{sum of rewards when } \mathit{a} \text{ taken prior to }\mathit{t}}{\text{number of times } \mathit{a} \text{ taken prior to }\mathit{t}} \\` `& = \frac{\displaystyle\sum_{i=1}^{t-1} R_i.\mathcal{1}_{A_i=a}}{\displaystyle\sum_{i=1}^{t-1} \mathcal{1}_{A_i=a}}` `\end{align}`
$A_t=\underset{a}{\mathrm{argmax}}{\text{ }Q_t(a)}$	`A_t=\underset{a}{\mathrm{argmax}}{\text{ }Q_t(a)}`
$p(s',r\|s,a) \doteq Pr\{S_t=s', R_t=r\|S_{t-1}=s, A_{t-1}=a\}$	`p(s',r\|s,a) \doteq Pr\{S_t=s', R_t=r\|S_{t-1}=s, A_{t-1}=a\}`
$q_\pi(s,a) \doteq \mathbb{E}[R_{t+1}+\gamma.G_{t+1}\|S_t=s, A_t=a]$	`q_\pi(s,a) \doteq \mathbb{E}[R_{t+1}+\gamma.G_{t+1}\|S_t=s, A_t=a]`
$v_*(s)\doteq \max\limits_{\pi} v_\pi(s), \forall s \in S$	`v_*(s)\doteq \max\limits_{\pi} v_\pi(s), \forall s \in S`
$q_\pi(s,a) \doteq \mathbb{E}[R_{t+1}+\gamma.G_{t+1}\|S_t=s, A_t=a]$	`q_\pi(s,a) \doteq \mathbb{E}[R_{t+1}+\gamma.G_{t+1}\|S_t=s, A_t=a]`
$l(w,b)=\frac{1}{N}\displaystyle\sum_{n=1}^{N}(y_n-(x_nw+b))^2$	`l(w,b)=\frac{1}{N}\displaystyle\sum_{n=1}^{N}(y_n-(x_nw+b))^2`
$\nabla l(w,b) = \begin{bmatrix}\frac{\partial l(w,b)}{\partial w_1}\\ \vdots \\\frac{\partial l(w,b)}{\partial w_d}\end{bmatrix}$	`\nabla l(w,b) = \begin{bmatrix}\frac{\partial l(w,b)}{\partial w_1}\\ \vdots \\\frac{\partial l(w,b)}{\partial w_d}\end{bmatrix}`
$\\ H(X) = – \sum_{x \in X} P(x) * \log(P(x))$	`\\ H(X) = – \sum_{x \in X} P(x) * \log(P(x))`
$X \sim \mathcal{N}(\mu,\,\sigma^{2})$	`X \sim \mathcal{N}(\mu,\,\sigma^{2})`
$\sqrt[n]{1+x+x^2+x^3+\dots+x^n}$	`\sqrt[n]{1+x+x^2+x^3+\dots+x^n}`

LaTeX	Markdow
\(\widehat{q}\)	`\widehat{q}`
\(\|x-x_M\|\)	`\|x-x_M\|`
\(f_M(x)\)	`f_M(x)`
\(\displaystyle\sum_{m}\)	`\displaystyle\sum_{m}`
\(\langle{x,w_m}\rangle\)	`\langle{x,w_m}\rangle`
\(\\| v_m \\|\)	`\\| v_m \\|`
\(\cos\)	`\cos`
\(\mathbb{Z}^d\)	`\mathbb{Z}$^d`
\(\infty\)	`\infty`
\(f \in L^2\)	`f \in L^2`
$$	`\implies`
\(\lim\limits_{M \to \infty}\)	`\lim\limits_{M \to \infty}`
\(P(A\|B) = \frac{P(B\|A)*P(A)}{P(B)}\)	`P(A\|B) = \frac{P(B\|A)*P(A)}{P(B)}`
\(\begin{align} \\ Q_t(a) &= \frac{\text{sum of rewards when } \mathit{a} \text{ taken prior to }\mathit{t}}{\text{number of times } \mathit{a} \text{ taken prior to }\mathit{t}} \\ & = \frac{\displaystyle\sum_{i=1}^{t-1} R_i.\mathcal{1}_{A_i=a}}{\displaystyle\sum_{i=1}^{t-1} \mathcal{1}_{A_i=a}} \end{align}\)	`\begin{align} \\` `Q_t(a) &= \frac{\text{sum of rewards when } \mathit{a} \text{ taken prior to }\mathit{t}}{\text{number of times } \mathit{a} \text{ taken prior to }\mathit{t}} \\` `& = \frac{\displaystyle\sum_{i=1}^{t-1} R_i.\mathcal{1}_{A_i=a}}{\displaystyle\sum_{i=1}^{t-1} \mathcal{1}_{A_i=a}}` `\end{align}`
\(A_t=\underset{a}{\mathrm{argmax}}{\text{ }Q_t(a)}\)	`A_t=\underset{a}{\mathrm{argmax}}{\text{ }Q_t(a)}`
\(p(s',r\|s,a) \doteq Pr\{S_t=s', R_t=r\|S_{t-1}=s, A_{t-1}=a\}\)	`p(s',r\|s,a) \doteq Pr\{S_t=s', R_t=r\|S_{t-1}=s, A_{t-1}=a\}`
\(q_\pi(s,a) \doteq \mathbb{E}[R_{t+1}+\gamma.G_{t+1}\|S_t=s, A_t=a]\)	`q_\pi(s,a) \doteq \mathbb{E}[R_{t+1}+\gamma.G_{t+1}\|S_t=s, A_t=a]`
\(v_*(s)\doteq \max\limits_{\pi} v_\pi(s), \forall s \in S\)	`v_*(s)\doteq \max\limits_{\pi} v_\pi(s), \forall s \in S`
\(q_\pi(s,a) \doteq \mathbb{E}[R_{t+1}+\gamma.G_{t+1}\|S_t=s, A_t=a]\)	`q_\pi(s,a) \doteq \mathbb{E}[R_{t+1}+\gamma.G_{t+1}\|S_t=s, A_t=a]`
\(l(w,b)=\frac{1}{N}\displaystyle\sum_{n=1}^{N}(y_n-(x_nw+b))^2\)	`l(w,b)=\frac{1}{N}\displaystyle\sum_{n=1}^{N}(y_n-(x_nw+b))^2`
\(\nabla l(w,b) = \begin{bmatrix}\frac{\partial l(w,b)}{\partial w_1}\\ \vdots \\\frac{\partial l(w,b)}{\partial w_d}\end{bmatrix}\)	`\nabla l(w,b) = \begin{bmatrix}\frac{\partial l(w,b)}{\partial w_1}\\ \vdots \\\frac{\partial l(w,b)}{\partial w_d}\end{bmatrix}`
\(\\ H(X) = – \sum_{x \in X} P(x) * \log(P(x))\)	`\\ H(X) = – \sum_{x \in X} P(x) * \log(P(x))`
\(X \sim \mathcal{N}(\mu,\,\sigma^{2})\)	`X \sim \mathcal{N}(\mu,\,\sigma^{2})`
\(\sqrt[n]{1+x+x^2+x^3+\dots+x^n}\)	`\sqrt[n]{1+x+x^2+x^3+\dots+x^n}`