How Does Meta RL Relate to Auto RL?

How does meta RL relate to say, auto RL? Are those two things related? So we were just talking about this. Auto RL is any way you can automate a piece of the RL pipeline. Data RL specifically is when you learn to do to replace a particular component in the RL algorithm. And I think auto RL isn't really a problem setting in that same sense. The meta RL problem setting is really central. That's kind of where most of this work comes from as well.