reinforcement learning1 [인공지능] Reinforcement Learning Summary ✨ Reinforcement Learning ◾ Any algorithms that solves MDP ◾ Learning to choose optimal action A → ㅠ* ◾ By learning evaluation functions like V(S), Q(S,A) ✨ Key ✔ next state function 을 알고있다면 ◾ use dynamic programming to learn ܸV(S) ◾ once learned, choose action At that maximizes V(S_t+1) ✔ 미로 찾기처럼 next state function 을 모른다면 ◾ Learn Q(St, At) = E[ V(S_t+1) ] ◾ 학습하기 위해 St x At → S_t+1 을 시도 .. 2022. 6. 14. 이전 1 다음 728x90
