from Hacker News

Challenges of real-world reinforcement learning

by motivic on 1/13/20, 4:44 PM with 5 comments

• by orasis on 1/13/20, 11:06 PM

  Reinforcement learning without simulation is not feasible for complex problems.

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

• by throwlaplace on 1/13/20, 8:10 PM

  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?