Interview with Safa Alver: Scalable and robust planning in lifelong reinforcement learning