Action-Sufficient State Representation Learning for Control with Structural Constraints

Open in new window