Metalearning generalizable dynamics from trajectories

no code implementations • 3 Jan 2023 • Qiaofeng Li, Tianyi Wang, Vwani Roychowdhury, M. Khalid Jawed

We present the interpretable meta neural ordinary differential equation (iMODE) method to rapidly learn generalizable (i. e., not parameter-specific) dynamics from trajectories of multiple dynamical systems that vary in their physical parameters.

Inductive Bias Meta-Learning