Corpus ID: 225067856

Simplifying Hamiltonian and Lagrangian Neural Networks via Explicit Constraints

@article{Finzi2020SimplifyingHA,
  title={Simplifying Hamiltonian and Lagrangian Neural Networks via Explicit Constraints},
  author={Marc Finzi and K. Wang and A. Wilson},
  journal={ArXiv},
  year={2020},
  volume={abs/2010.13581}
}
  • Marc Finzi, K. Wang, A. Wilson
  • Published 2020
  • Computer Science, Mathematics, Physics
  • ArXiv
  • Reasoning about the physical world requires models that are endowed with the right inductive biases to learn the underlying dynamics. Recent works improve generalization for predicting trajectories by learning the Hamiltonian or Lagrangian of a system rather than the differential equations directly. While these methods encode the constraints of the systems using generalized coordinates, we show that embedding the system into Cartesian coordinates and enforcing the constraints explicitly with… CONTINUE READING
    3 Citations
    Structure Preservation for the Deep Neural Network Multigrid Solver
    • PDF
    Benchmarking Energy-Conserving Neural Networks for Learning Dynamics from Data
    • 1
    • Highly Influenced
    • PDF

    References

    SHOWING 1-10 OF 24 REFERENCES
    Lagrangian Neural Networks
    • 40
    • PDF
    Symplectic ODE-Net: Learning Hamiltonian Dynamics with Control
    • 43
    • Highly Influential
    • PDF
    Hamiltonian Neural Networks
    • 125
    • Highly Influential
    • PDF
    Symplectic Recurrent Neural Networks
    • 39
    • Highly Influential
    • PDF
    Deep Hamiltonian networks based on symplectic integrators
    • 5
    • PDF
    Hamiltonian Graph Networks with ODE Integrators
    • 38
    • PDF
    Deep Lagrangian Networks: Using Physics as Model Prior for Deep Learning
    • 75
    • Highly Influential
    • PDF
    Dissipative SymODEN: Encoding Hamiltonian Dynamics with Dissipation and Control into Deep Learning
    • 8
    • PDF
    Neural Ordinary Differential Equations
    • 826
    • PDF