Almost Surely Stable Deep Dynamics
Abstract
We introduce a method for learning provably stable deep neural network based dynamic models from observed data. Specifically, we consider discrete-time stochastic dynamic models, as they are of particular interest in practical applications such as estimation and control. However, these aspects exacerbate the challenge of guaranteeing stability. Our method works by embedding a Lyapunov neural network into the dynamic model, thereby inherently satisfying the stability criterion. To this end, we propose two approaches and apply them in both the deterministic and stochastic settings: one exploits convexity of the Lyapunov function, while the other enforces stability through an implicit output layer. We demonstrate the utility of each approach through numerical examples.
Cite
@article{arxiv.2103.14722,
title = {Almost Surely Stable Deep Dynamics},
author = {Nathan P. Lawrence and Philip D. Loewen and Michael G. Forbes and Johan U. Backström and R. Bhushan Gopaluni},
journal= {arXiv preprint arXiv:2103.14722},
year = {2021}
}
Comments
NeurIPS 2020; Spotlight Paper