Gradient-free training of neural ODEs for system identification and control using ensemble Kalman inversion
Keywords: dynamical systems, data-driven dynamics, neural ODEs, optimal control, ensemble Kalman inversion
TL;DR: This work shows that Ensemble Kalman inversion (EKI) is an efficient gradient-free optimization method for training neural neural ODEs in system identification and optimal control tasks.
Abstract: Ensemble Kalman inversion (EKI) is a sequential Monte Carlo method used to solve inverse problems within a Bayesian framework. Unlike backpropagation, EKI is a gradient-free optimization method that only necessitates the evaluation of artificial neural networks in forward passes. In this study, we examine the effectiveness of EKI in training neural ordinary differential equations (neural ODEs) for system identification and control tasks. To apply EKI to optimal control problems, we formulate inverse problems that incorporate a Tikhonov-type regularization term. Our numerical results demonstrate that EKI is an efficient method for training neural ODEs in system identification and optimal control problems, with runtime and quality of solutions that are competitive with commonly used gradient-based optimizers.
Submission Number: 2