> I don't think ACT (unlike MPC) have an explicit use case for different embodiment. Or what do you mean by two different instances? I mean two different Aloha

> I don't think ACT (unlike MPC) have an explicit use case for different embodiment. Or what do you mean by two different instances? I mean two different Aloha setups (2 x 2x WidowX) in the same lab. This is the very minimal level of "generalization" you may expect, barely enough to be called generalization, but still, the learned policy seems to overfit to the individual robot's joint space quirks. > in the field of BC (Not IRL) what have been the more robust method for your case? Different things for different definitions of robust. What's your definition?