Search on the Replay Buffer:
        Bridging Planning and Reinforcement Learning

         Benjamin Eysenbachθφ , Ruslan Salakhutdinovθ , Sergey Levineφψ
               θ
                CMU, φ Google Brain, ψ UC Berkeley
                 beysenba@cs.cmu.edu

                        Abstract
      The history of learning for control has been an exciting back and forth between
      two broad classes of algorithms: planning and reinforcement learning. Planning
      algorithms effectively reason ...