Introducing XIRL, a self-supervised method for Cross-embodiment Inverse RL, which summarizes task objective knowledge from videos in the form of reward functions used to teach tasks to robots with new physical embodiments. Read more and copy the code ↓ https://t.co/9WjjM72Tc4
@GoddenThomas I understand now, thank you. With similar math it moves target to the actual balance point, not try to move balance point to slightly incorrect target (which seems to be impossible).
@arturo182@zuph That is why I always check the datasheet and pay attention how chip is oriented in packaging and with regards to signs on it. Helps to disambiguate.