Wait a moment...
Exploring the intersection of Code, Data, and Life.
1개의 글
AI#Advantage
DeepMind의 A3C 논문 「Asynchronous Methods for Deep Reinforcement Learning」을 바탕으로 asynchronous learning, Actor-Critic, Advantage, n-step return, entropy regularization을 DDQN과 비교하며 정리한다.