Tweeted By @DeepMind
How can RL be made usable in the real world? Offline RL is part of the solution but we need to pick hyperparameters using offline data too.
— DeepMind (@DeepMind) August 4, 2020
Researchers show that certain simple approaches can be remarkably effective at offline hyperparameter selection: https://t.co/XsxFJXDi8y pic.twitter.com/piZdxEgDSi