When to use parametric models
           in reinforcement learning?

   Hado van Hasselt        Matteo Hessel          John Aslanides
     DeepMind           DeepMind             DeepMind
     London, UK          London, UK            London, UK
   hado@google.com       mtthss@google.com        jaslanides@google.com

                       Abstract
     We examine the question of when and how parametric models are most useful in
     reinforcement learning. In particular, we look at commonalitie ...