Sparse Symplectically Integrated Neural Networks
arXiv:2006.12972
Abstract
We introduce Sparse Symplectically Integrated Neural Networks (SSINNs), a novel model for learning Hamiltonian dynamical systems from data. SSINNs combine fourth-order symplectic integration with a learned parameterization of the Hamiltonian obtained using sparse regression through a mathematically elegant function space. This allows for interpretable models that incorporate symplectic inductive biases and have low memory requirements. We evaluate SSINNs on four classical Hamiltonian dynamical problems: the Hénon-Heiles system, nonlinearly coupled oscillators, a multi-particle mass-spring system, and a pendulum system. Our results demonstrate promise in both system prediction and conservation of energy, often outperforming the current state-of-the-art black-box prediction techniques by an order of magnitude. Further, SSINNs successfully converge to true governing equations from highly limited and noisy data, demonstrating potential applicability in the discovery of new physical governing equations.
arXiv commentAccepted as a conference paper to NeurIPS 2020. Main paper has 9 pages and 4 figures
References in corpus (13)
- Discovering governing equations from data: Sparse identification of nonlinear dynamical systems
- Hidden Physics Models: Machine Learning of Nonlinear Partial Differential Equations
- Data-driven discovery of coordinates and governing equations
- Discovering physical concepts with neural networks
- Hamiltonian Neural Networks
- Attractor Reconstruction by Machine Learning
- Data Driven Governing Equations Approximation Using Deep Neural Networks
- Discovering Symbolic Models from Deep Learning with Inductive Biases
- Plug and Play Language Models: A Simple Approach to Controlled Text Generation
- Newton vs the machine: solving the chaotic three-body problem using deep neural networks
- Applying machine learning to improve simulations of a chaotic dynamical system using empirical error correction
- Symplectic Recurrent Neural Networks
- Variational Integrator Networks for Physically Structured Embeddings