2018年5月26日
CS294-112 深度强化学习 秋季学期（伯克利）NO.7 Optimal control and planning
摘要： transition possibility is unknown and we even don't need to estimate the possibility 阅读全文
posted @ 2018-05-26 23:04 ecoflex 阅读(146) 评论(0) 推荐(0)
CS294-112 深度强化学习 秋季学期（伯克利）NO.6 Value functions introduction NO.7 Advanced Q learning
摘要： understand that correlated samples cause problem. and how paralled solve the problem another solution is replay buffers, fully ultilizing the advantag 阅读全文
posted @ 2018-05-26 19:57 ecoflex 阅读(220) 评论(0) 推荐(0)
CS294-112 深度强化学习 秋季学期（伯克利）NO.5 Actor-critic introduction
摘要： in most AC algorithms, we actually just fit value function. less common to fit Q function as well. batch：off line， monte carlo。online： bootstrap，TD in 阅读全文
posted @ 2018-05-26 12:28 ecoflex 阅读(213) 评论(0) 推荐(0)