RRL: Resnet as representation for Reinforcement Learning

                       Rutav Shah * 1 Vikash Kumar * 2

             Abstract
   The ability to autonomously learn behaviors via                         Supervised Learning
   direct interactions in uninstrumented environ-
   ments can lead to generalist robots capable of en-
   hancing productivity or providing care in unstruc-
   tured settings like homes. Such uninstrumented                 Reinforcement
                                ...