Algorithms - Equations

Question 1

V(s) = max_a ( R(s,a) + γΣ_S1 T(s,a,s1) V(s1))

Answer 1

Bellman Equation

The value of a state equals the max over all the actions - the reward you get for taking that action in that state plus the discounted value of state you end up in weighted by the probability you end up there.

Question 2

Explain the belman equation.

Answer 2

The value or utility of a state is based on the discounted reward of future rewards from state, action, state prime until terminal.