D-DL4CV-Lec21c-Q-Function

Q-Function

The Q-function at state $s$ doing action $a$ computes the expected reward from following policy $\pi$

$$Q^{\pi }(s,a)=\mathbb{E}\left[\sum _{t\ge 0}{\gamma }^t{r}_t|s_0=s,a_0=a,\pi \right]$$