MDP Specifications in the Real World

$0$ points possible $($ungraded$)$

We would need to provide the following tuple for our RL algorithms in an MDP setting.

However, in the real world, not all of these might be directly available.

From the options below select, which one$($s$)$ would not be readily available in most real$-$world scenarios.

S: State Space

A: Action Space

T: Transition Probabilities

R: Reward Structure