machinelearning icon indicating copy to clipboard operation
machinelearning copied to clipboard

关于更新values

Open jangXiaoFan opened this issue 4 years ago • 0 comments

强化学习第一篇,第218行,更新estimations时,为什么要过滤掉探索动作的收益,这样的话探索率epsilon还有意义吗?

jangXiaoFan avatar Oct 22 '20 10:10 jangXiaoFan