Learning Poisson systems and trajectories of autonomous systems via Poisson neural networks

We propose the Poisson neural networks (PNNs) to learn Poisson systems and trajectories of autonomous systems from data. Based on the Darboux-Lie theorem, the phase flow of a Poisson system can be written as the composition of (1) a coordinate transformation, (2) an extended symplectic map and (3) the inverse of the transformation. In this work, we extend this result to the unknotted trajectories of autonomous systems. We employ structured neural networks with physical priors to approximate the three aforementioned maps. We demonstrate through several simulations that PNNs are capable of handling very accurately several challenging tasks, including the motion of a particle in the electromagnetic potential, the nonlinear Schr{\"o}dinger equation, and pixel observations of the two-body problem.

[1]  Ioannis G. Kevrekidis,et al.  Identification of distributed parameter systems: A neural net based approach , 1998 .

[2]  Samy Bengio,et al.  Density estimation using Real NVP , 2016, ICLR.

[3]  E. Hairer,et al.  Geometric Numerical Integration: Structure Preserving Algorithms for Ordinary Differential Equations , 2004 .

[4]  Andrew Gordon Wilson,et al.  Simplifying Hamiltonian and Lagrangian Neural Networks via Explicit Constraints , 2020, NeurIPS.

[5]  Aiqing Zhu,et al.  Unit Triangular Factorization of the Matrix Symplectic Group , 2019, SIAM J. Matrix Anal. Appl..

[6]  David Duvenaud,et al.  Neural Ordinary Differential Equations , 2018, NeurIPS.

[7]  Kyle Cranmer,et al.  Hamiltonian Graph Networks with ODE Integrators , 2019, ArXiv.

[8]  I.G. Kevrekidis,et al.  Continuous-time nonlinear signal processing: a neural network based approach for gray box identification , 1994, Proceedings of IEEE Workshop on Neural Networks for Signal Processing.

[9]  Jason Yosinski,et al.  Hamiltonian Neural Networks , 2019, NeurIPS.

[10]  Ioannis G. Kevrekidis,et al.  A comparison of recurrent training algorithms for time series analysis and system identification , 1996 .

[11]  R. A. Silverman,et al.  The Mathematical Theory of Viscous Incompressible Flow , 2014 .

[12]  Amit Chakraborty,et al.  Symplectic ODE-Net: Learning Hamiltonian Dynamics with Control , 2020, ICLR.

[13]  M. Kramer.  Nonlinear principal component analysis using autoassociative neural networks , 1991 .

[14]  Danilo Jimenez Rezende,et al.  Equivariant Hamiltonian Flows , 2019, ArXiv.

[15]  George Em Karniadakis,et al.  SympNets: Intrinsic structure-preserving symplectic networks for identifying Hamiltonian systems , 2020, Neural Networks.

[16]  Víctor M. Pérez-García,et al.  Symplectic methods for the nonlinear Schrödinger equation , 1996 .

[17]  Andrew Jaegle,et al.  Hamiltonian Generative Networks , 2020, ICLR.

[18]  On Difference Schemes and Symplectic Geometry ? X1 Introductory Remarks , 2022 .

[19]  C. Lubich.  From Quantum to Classical Molecular Dynamics: Reduced Models and Numerical Analysis , 2008 .

[20]  Jianfeng Lu,et al.  A Mean-field Analysis of Deep ResNet and Beyond: Towards Provable Optimization Via Overparameterization From Depth , 2020, ICML.

[21]  Yifa Tang,et al.  Inverse modified differential equations for discovery of dynamics , 2020, ArXiv.

[22]  Ioannis G. Kevrekidis,et al.  DISCRETE- vs. CONTINUOUS-TIME NONLINEAR SIGNAL PROCESSING OF Cu ELECTRODISSOLUTION DATA , 1992 .

[23]  E Weinan,et al.  A Proposal on Machine Learning via Dynamical Systems , 2017, Communications in Mathematics and Statistics.

[24]  Eldad Haber,et al.  Reversible Architectures for Arbitrarily Deep Residual Neural Networks , 2017, AAAI.

[25]  G. Karniadakis,et al.  Multistep Neural Networks for Data-driven Discovery of Nonlinear Dynamical Systems , 2018, 1801.01236.

[26]  A. Blumberg.  BASIC TOPOLOGY , 2002 .

[27]  Jaideep Pathak,et al.  Backpropagation algorithms and Reservoir Computing in Recurrent Neural Networks for the forecasting of complex spatiotemporal dynamics , 2019, Neural Networks.

[28]  Yoshua Bengio,et al.  NICE: Non-linear Independent Components Estimation , 2014, ICLR.

[29]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[30]  Miles Cranmer,et al.  Lagrangian Neural Networks , 2020, ICLR 2020.

[31]  Eldad Haber,et al.  Stable architectures for deep neural networks , 2017, ArXiv.

[32]  T. V. H. Prathamesh.  Knot Theory , 2016, Arch. Formal Proofs.

[33]  Jianyu Zhang,et al.  Symplectic Recurrent Neural Networks , 2020, ICLR.

[34]  Ioannis G. Kevrekidis,et al.  On learning Hamiltonian systems from data. , 2019, Chaos.

[35]  Bin Dong,et al.  Beyond Finite Layer Neural Networks: Bridging Deep Architectures and Numerical Differential Equations , 2017, ICML.

[36]  Andrew Ranicki,et al.  High-dimensional Knot Theory: Algebraic Surgery in Codimension 2 , 2010 .

[37]  Cheng Yang,et al.  Nonseparable Symplectic Neural Networks , 2021, ICLR.