In this work, we propose a BNN-based dynamics learning method using physical priors for efficient policy learning.
Jan 8, 2025