Off-policy is the new on-policy
Sarah Catanzaro says off-policy methods are replacing on-policy RL
The shift allows developers to reuse older training data.
219002.5K
Original post unavailable.
Sentiment
Users are praising Richard Sutton's repeated contributions as off-policy reinforcement learning emerges as the preferred approach.
Pos
100.0%
Neg
0.0%
1 comments with sentiment.
Cluster Engagement
Digg Deeper
No Digg Deeper questions have been answered for this story yet.
Posts from X
Most Activity
Most Activity
VIEWS2.5KLIKES19REPLIES2
@sarahcat21 Some have been saying…
Off-policy is the new on-policy
@sarahcat21 Richard Sutton strikes again