SupplementarymaterialforVariationalAutomatic CurriculumLearningforSparse-Reward CooperativeMulti-AgentProblems