Inverse modified differential equations for discovery of dynamics

09/02/2020
by   Aiqing Zhu, et al.
0

We introduce inverse modified differential equations (IMDEs) to contribute to the fundamental theory of discovery of dynamics. In particular, we investigate the IMDEs for the neural ordinary differential equations (neural ODEs). Training such a learning model actually returns an approximation of an IMDE, rather than the original system. Thus, the convergence analysis for data-driven discovery is illuminated. The discrepancy of discovery depends on the order of the integrator used. Furthermore, IMDEs make clear the behavior of parameterizing some blocks in neural ODEs. We also perform several experiments to numerically substantiate our theoretical results.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 1

page 2

page 3

page 4

12/19/2019

Polynomial Neural Networks and Taylor maps for Dynamical Systems Simulation and Learning

The connection of Taylor maps and polynomial neural networks (PNN) to so...
12/29/2019

Discovery of Dynamics Using Linear Multistep Methods

Linear multistep methods (LMMs) are popular time discretization techniqu...
11/07/2020

Latent Neural Differential Equations for Video Generation

Generative Adversarial Networks have recently shown promise for video ge...
07/23/2021

Forecasting the outcome of spintronic experiments with Neural Ordinary Differential Equations

Deep learning has an increasing impact to assist research, allowing, for...
01/13/2021

Neuro-Reachability of Networked Microgrids

A neural ordinary differential equations network (ODE-Net)-enabled reach...
01/02/2018

Computing the Inverse Mellin Transform of Holonomic Sequences using Kovacic's Algorithm

We describe how the extension of a solver for linear differential equati...
02/05/2002

A Qualitative Dynamical Modelling Approach to Capital Accumulation in Unregulated Fisheries

Capital accumulation has been a major issue in fisheries economics over ...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.