We provide a novel characterization of augmented balancing weights, also...
Offline reinforcement learning is important in domains such as medicine,...
When decision-makers can directly intervene, policy evaluation algorithm...
We study balancing weight estimators, which reweight outcomes from a sou...
share