AWR learns policies via advantage-weighted supervised regression on actions, achieving competitive off-policy performance on Gym tasks and strong results from static data alone.
URL http://proceedings.mlr.press/v80/fujimoto18a.html
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Advantage-Weighted Regression: Simple and Scalable Off-Policy Reinforcement Learning
AWR learns policies via advantage-weighted supervised regression on actions, achieving competitive off-policy performance on Gym tasks and strong results from static data alone.