Deep Hamiltonian networks based on symplectic integrators

by   Aiqing Zhu, et al.

HNets is a class of neural networks on grounds of physical prior for learning Hamiltonian systems. This paper explains the influences of different integrators as hyper-parameters on the HNets through error analysis. If we define the network target as the map with zero empirical loss on arbitrary training data, then the non-symplectic integrators cannot guarantee the existence of the network targets of HNets. We introduce the inverse modified equations for HNets and prove that the HNets based on symplectic integrators possess network targets and the differences between the network targets and the original Hamiltonians depend on the accuracy orders of the integrators. Our numerical experiments show that the phase flows of the Hamiltonian systems obtained by symplectic HNets do not exactly preserve the original Hamiltonians, but preserve the network targets calculated; the loss of the network target for the training data and the test data is much less than the loss of the original Hamiltonian; the symplectic HNets have more powerful generalization ability and higher accuracy than the non-symplectic HNets in addressing predicting issues. Thus, the symplectic integrators are of critical importance for HNets.


page 1

page 2

page 3

page 4


Symplectic Learning for Hamiltonian Neural Networks

Machine learning methods are widely used in the natural sciences to mode...

Port-Hamiltonian Systems Modelling in Electrical Engineering

The port-Hamiltonian modelling framework allows for models that preserve...

Symplectic networks: Intrinsic structure-preserving networks for identifying Hamiltonian systems

This work presents a framework of constructing the neural networks prese...

Preservation of Quadratic Invariants by Semiexplicit Symplectic Integrators for Non-separable Hamiltonian Systems

We prove that the recently developed semiexplicit symplectic integrators...

Time Discretizations of Wasserstein-Hamiltonian Flows

We study discretizations of Hamiltonian systems on the probability densi...

Data-driven Prediction of General Hamiltonian Dynamics via Learning Exactly-Symplectic Maps

We consider the learning and prediction of nonlinear time series generat...

Please sign up or login with your details

Forgot password? Click here to reset