Introducing TayPO, a unifying framework that generalises prior work as first order special cases, draws close connections between off-policy evaluation & policy optimisation, and brings empirical gains on a few distributed deep RL agents.https://t.co/csEWLTxS2Y #ICML2020 pic.twitter.com/guI2aQkyc4
— DeepMind (@DeepMind) July 13, 2020
from Twitter https://twitter.com/DeepMind
July 13, 2020 at 06:28AM
via IFTTT
No comments:
Post a Comment