REVIEW 1 cited by
Efficient iterative policy optimization
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
We tackle the issue of finding a good policy when the number of policy updates is limited. This is done by approximating the expected policy reward as a sequence of concave lower bounds which can be efficiently maximized, drastically reducing the number of policy updates required to achieve good performance. We also extend existing methods to negative rewards, enabling the use of control variates.
Forward citations
Cited by 1 Pith paper
-
Supervised Fine Tuning on Curated Data is Reinforcement Learning (and can be improved)
SFT on curated data is a lower bound on a sparse-reward RL objective, and an importance-weighted variant, iw-SFT, tightens the bound and beats plain SFT on AIME 2024 and GPQA.
Discussion (0). Continue with ORCID to comment.