Reinforcement Learning,

https://habr.com/ru/post/437020/
https://www.alexirpan.com/2018/02/14/rl-hard.html

DRL Карпатый

http://karpathy.github.io/2016/05/31/rl/