Challenges of real-world reinforcement learning

48 points by motivic 6 years ago | 5 comments

orasis 6 years ago |

Reinforcement learning without simulation is not feasible for complex problems.

If you can’t simulate, stick to multi-armed bandits.

anyone know a good survey paper about current SOA? i've read sutton&barton, skimmed openai "Spinning Up in Deep RL". anything else out there?

hcrisp 6 years ago | |

The Deep RL Bootcamp lectures are still good and relevant:

throwlaplace 6 years ago | | |

beggars can't be choosers of course but i would prefer something to read rather than something to watch

orasis 6 years ago | |