DEPO uses historical data to build a data-dependent uncertainty bonus for exploration in online RLHF, yielding an adaptive regret bound and stronger empirical performance than baselines.
arXiv preprint arXiv:2502.07193 , year=
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Data-dependent Exploration for Online Reinforcement Learning from Human Feedback
DEPO uses historical data to build a data-dependent uncertainty bonus for exploration in online RLHF, yielding an adaptive regret bound and stronger empirical performance than baselines.