Estimating the expected return under a specified control policy. It is distinct from improving or optimizing that policy. A Bellman expectation equation describes the fixed-policy value, whereas a Bellman optimality equation compares admissible actions.
New to topics? Read the docs here!