New paper - CURL: Contrastive Unsupervised Representations for RL! We use the simplest form of contrastive learning (instance-based) as an auxiliary task in model-free RL. SoTA by *significant* margin on DMControl and Atari for data-efficiency. https://t.co/Xszek8cFFs pic.twitter.com/3ZSfzmxemE
— Aravind Srinivas (@Aravind7694) April 9, 2020
from Twitter https://twitter.com/Aravind7694
April 08, 2020 at 05:46PM
via IFTTT
No comments:
Post a Comment