Offline policy optimization under all-policy linear realizability
The method uses one convex regression per layer and has sample dependence of order epsilon to the minus two.
Stories per day · 30 days
1 RSS
Relatedreinforcement-learning3causal-inference1constrained-mdp1data-selection1diffusion-policy1next-best-action1ogbench1policy-optimization1q-learning1rideshare1
8 stories, newest first · one card per story
Stories tagged #offline-rl
HQL balances one-step and multi-step returns for long horizons
A multi-step target is kept only when a dataset action beats best-of-N behavior-cloned actions.
GIST improves offline RLVR accuracy by selective token supervision
GIST approximates Gibbs-optimal policy to select informative responses and tokens, outperforming baselines on GSM8K MATH-500 MBPP while supervising 66.6% fewer tokens.
Offline RL frames next-best action as a constrained decision process
In a rideshare coupon study, the authors report 19% higher lifetime value than the behavior policy.
Retimed Bellman Flows: Escaping the Impossible Triangle of Velocity Bootstrapping
Retimed Bellman Flows cut return-distribution error up to 7.7× and beat prior flow critics on 38 offline RL tasks.
Trust Guided Decision Transformer
Trust Guided Decision Transformer filters drifted context with prediction error before applying value guidance.
JumpStart Your Policy Learning with Lessons from 160,000 Training Runs
A study of 160,000 offline RL and imitation policies finds no dominant algorithm and releases JumpStart.
Learning Multimodal One-step Flow Policy via Value-weighted Optimal Transport
OptiFlow learns one-step multimodal flow policies for offline RL via state-wise entropic optimal transport, avoiding critic overestimation and mode collapse.