RL-Ch8-Imitation Learning 2020-10-28 RL-Ch8-Imitation Learning要使Agent有人的智慧有两种手段: 强化学习(Reinforcement Learning) 模仿学习(Imitation Learning) more >> Reinforcement Learning computer 展开全文 >>
RL-Ch7-Reward Issue 2020-10-27 RL-Ch7-Reward Issue因为采样的不充分性和种种原因(环境v很少给reward),我们得到环境的回馈信息是很少的,这就是奖励稀疏(sparse reward)的问题。 more >> Reinforcement Learning computer 展开全文 >>
RL-Ch6-Actor-Critic 2020-10-26 RL-Ch6-Actor-CriticA2C:Advantage Actor-Critic A3C:Asynchronous Actor-Critic more >> Reinforcement Learning computer 展开全文 >>
RL-Ch5-Q-Learning 2020-10-25 RL-Ch5-Q-Learning本文提到的actor等效于前几章笔记中的agent+policy。 more >> Reinforcement Learning computer 展开全文 >>
RL-Ch4-Policy Gradient 2020-10-24 RL-Ch4-Policy Gradient策略梯度(Policy Gradient) more >> Reinforcement Learning computer 展开全文 >>
RL-Ch3-基于表格方法求解RL 2020-10-23 RL-Ch3-基于表格方法求解RL 本文不太完善,待寻到时间再细细修改。 more >> Reinforcement Learning computer 展开全文 >>
RL表格型方法求解——Q-learning 2020-10-23 RL表格型方法求解—— Q-learning参考来自百度强化学习课程 more >> Reinforcement Learning computer 展开全文 >>
RL表格型方法求解——Sarsa 2020-10-23 RL表格型方法求解——Sarsa参考来自百度强化学习课程 more >> Reinforcement Learning computer 展开全文 >>
RL-Ch2-马尔可夫决策过程 2020-10-22 RL-Ch2-马尔可夫决策过程前情回顾RL Agent的核心元素为model,value,policy。 more >> Reinforcement Learning computer 展开全文 >>