Technology & AIAug 25, 2026
Bellman Calibration for Marginalized Importance Weighting in Offline Reinforcement Learning
Marginalized importance weighting evaluates a target policy by reweighting offline state-action samples with its discounted occupancy ratio, characterized by an adjoint Bellman equation.
Marginalized importance weighting evaluates a target policy by reweighting offline state-action samples with its discounted occupancy ratio, characterized by an adjoint Bellman equation. Existing minimax, primal-dual, and fitted fixed-point estimators can leave residual…
Sign in to learn & save →
The frontier is open to all. Sign in to learn this from first principles and save it to your knowledge base.