System Identification Through Lipschitz Regularized Deep Neural Networks

Abstract

In this paper we use neural networks to learn governing equations from data. Specifically we reconstruct the right-hand side of a system of ODEs $\dot{x}(t)=f(t,x(t))$ directly from observed uniformly time-sampled data using a neural network. In contrast with other neural network based approaches to this problem, we add a Lipschitz regularization term to our loss function. In the synthetic examples we observed empirically that this regularization results in a smoother approximating function and better generalization properties when compared with non-regularized models, both on trajectory and non-trajectory data, especially in presence of noise. In contrast with sparse regression approaches, since neural networks are universal approximators, we don't need any prior knowledge on the ODE system. Since the model is applied component wise, it can handle systems of any dimension, making it usable for real-world data.

References in corpus (6)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Hidden Physics Models: Machine Learning of Nonlinear Partial Differential Equations
• Applied Koopmanism
• Spectrally-normalized margin bounds for neural networks
• Training robust neural networks using Lipschitz bounds
• Nonlinear Systems Identification Using Deep Dynamic Neural Networks

Cited by in corpus (4)

• A numerically efficient output-only system-identification framework for stochastically forced self-sustained oscillators
• Optimal Transport for Parameter Identification of Chaotic Dynamics via Invariant Measures
• Learning dynamics on invariant measures using PDE-constrained optimization
• A Neural Network Ensemble Approach to System Identification