VIREL :    A Variational Inference Framework
             for Reinforcement Learning

      Matthew Fellows Anuj Mahajan Tim G. J. Rudner         Shimon Whiteson
               Department of Computer Science
                 University of Oxford
                        Abstract
      Applying probabilistic models to reinforcement learning (RL) enables the uses
      of powerful optimisation tools such as variational inference in RL. However, ex-
      isting inference frameworks and their algo ...