trends.barnas.net

task · first seen 7 Nov 2026

safe offline reinforcement learning

Also: safe offline RL

Learning constrained policies from a static dataset so that the policy maximizes return while satisfying safety constraints.

Mentions per week

0119 May11 Aug10 NovPapersCodeDiscussion

PapersCodeDiscussion

Earliest sources

Latest sources

Mentioned together with

conditional variational autoencoderadvantage-weighted regression