MAMBPO: Sample-efficient multi-robot reinforcement learning using learned world models.

Danil Willemsen, Mario Coppola, Guido C. H. E. de Croon

Browse the full IROS paper archive.