trends.barnas.net

method · first seen 5 Oct 2026

offline-to-online reinforcement learning

Also: O2O RL, offline-to-online RL, offline data for online learning

Training a policy first on fixed logged data and then fine-tuning it with online interaction.

Mentions per week

0319 May11 Aug10 NovPapersCodeDiscussion

PapersCodeDiscussion

Earliest sources

Latest sources

Mentioned together with

imitation learningreward shapingdiffusion transformer cachinglinear banditdexterous manipulationsafe reinforcement learningreward modelingflow Q-learningKL regularizationoffline imitation learning