Closed joakimpersson closed 10 years ago
That \Pr(s_{t+1} | s_t, a_t, ... s_1, a1) = \Pr(s{t+1} | s_t, a_t) = P(s_t, at, s{t+1})
This should actually be stated in the definition of the MDP beforehand
That \Pr(s_{t+1} | s_t, a_t, ... s_1, a1) = \Pr(s{t+1} | s_t, a_t) = P(s_t, at, s{t+1})
This should actually be stated in the definition of the MDP beforehand