Re:Algorithms for Reinforcement Learning
by currymj on Monday February 18, 2019 @11:44PM
(#19196657)
Yes especially if you are willing to broaden your definition to include multi-armed or contextual bandits (of which Markov decision processes that RL algorithms solve are a generalization). Lot of internet ad companies etc. use those successfully.
[ Parent
| on HN ]
There are no comments on this comment yet.