2
$\begingroup$

I was reading a book on reinforcement learning, and came across the following notation:

enter image description here

Possibly stupid question, but I cant seem to google how to read this. Is it

A) Probability of (s' and r) GIVEN (s and a)

B) Probability of s' and (r GIVEN s) and a?

I know this may seem like a stupid question, but I really could not find any guide on how to read this triple notation online. Thank you for your time!

$\endgroup$
5
$\begingroup$

It is A.

It is the probability that the state ($S_t$) and reward ($R_t$) at time $t$ are respectively equal to some $s'$ and $r$, given that the state ($S_{t-1}$) and the action taken from that state ($A_{t-1}$) at time $t-1$ are respectively equal to some $s$ and $a$.

In other words, this is a "joint" probability where you are conditioning on more than one variable.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.