Self-Imitation Learning via Generalized Lower
           Bound Q-learning

                      Yunhao Tang
                     Columbia University
                    yt2541@columbia.edu

                       Abstract
     Self-imitation learning motivated by lower-bound Q-learning is a novel and effec-
     tive approach for off-policy learning. In this work, we propose a n-step lower bound
     which generalizes the original return-based lower-bound Q-learning, and introduce
     a new fa ...