What Did You Think Would Happen? Explaining
     Agent Behaviour through Intended Outcomes

     Herman Yau               Chris Russell          Simon Hadfield
  CVSSP, University of Surrey        Amazon Web Services      CVSSP, University of Surrey

                          Abstract
       We present a novel form of explanation for Reinforcement Learning, based around
       the notion of intended outcome. These explanations describe the outcome an agent
       is trying to achieve by its act ...