Social Reinforcement Learning talk at RLDM