Towards a more efficient computation of individual attribute and policy contribution for post-hoc explanation of cooperative multi-agent systems using Myerson values
Giorgio Angelotti, Natalia Díaz-Rodríguez
Topics & Concepts
Computer scienceExploitA priori and a posterioriGraphReinforcement learningTask (project management)ComputationCrowdsourcingArtificial intelligenceShapley valueTheoretical computer scienceMachine learningGame theoryMathematical economicsMathematicsComputer securityAlgorithmWorld Wide WebManagementEpistemologyEconomicsPhilosophyReinforcement Learning in RoboticsExplainable Artificial Intelligence (XAI)Sports Analytics and Performance