luckydush / a05_rl Goto Github PK
View Code? Open in Web Editor NEWThis project forked from cv-xueba/a05_rl
本课程主要介绍强化学习的基础知识,其目标是帮助同学们快速、顺利地进入强化学习及其应用领域的研究工作。课程主要内容包含有限马尔可夫决策过程,动态规划,无模型预测与控制(SASA,Q-Learning),价值函数逼近(DQN),策略梯度方法(REINFORCE),执行者/评论者方法(AC,TRPO,PPO),连续动作空间的确定性策略(DDPG)。