Authors

Gordei Pribõtkin, Stefania Tomasiello

Abstract

In this short note, the performance of two kinds of physics-guided computing schemes, namely the Hamiltonian Neural Network and the Port-Hamiltonian Neural Network, are discussed through the predicted dynamics of two coupled Duffing oscillators. First, we propose a new error bound which holds for both types of networks. Then, we numerically investigate some alternative activation functions in terms of prediction accuracy. The numerical results show the potential of the approaches when compared to the standard neural networks in the transient regime.

Keywords

Port-Hamiltonian Neural Network; Hamiltonian Neural Network; Forced system; Damped system

Citation

  • Journal: Neural Processing Letters
  • Year: 2023
  • Volume: 55
  • Issue: 6
  • Pages: 8163–8180
  • Publisher: Springer Science and Business Media LLC
  • DOI: 10.1007/s11063-023-11306-0

BibTeX

@article{Prib_tkin_2023,
  title={{Using Hamiltonian Neural Networks to Model Two Coupled Duffing Oscillators}},
  volume={55},
  ISSN={1573-773X},
  DOI={10.1007/s11063-023-11306-0},
  number={6},
  journal={Neural Processing Letters},
  publisher={Springer Science and Business Media LLC},
  author={Pribõtkin, Gordei and Tomasiello, Stefania},
  year={2023},
  pages={8163--8180}
}

Download the bib file

References

  • SH Strogatz. Strogatz SH (2000) Nonlinear dynamics and chaos: with applications to physics, biology. Westview Press, Chemistry and Engineering (2000)
  • Awrejcewicz, J. Bifurcation and Chaos in Coupled Oscillators. (1991) doi:10.1142/1342 – 10.1142/1342
  • Deco, G., Kringelbach, M. L., Jirsa, V. K. & Ritter, P. The dynamics of resting fluctuations in the brain: metastability and its dynamical cortical core. Scientific Reports vol. 7 (2017) – 10.1038/s41598-017-03073-5
  • Qing, Y. M., Ren, Y., Lei, D., Ma, H. F. & Cui, T. J. Strong coupling in two-dimensional materials-based nanostructures: a review. Journal of Optics vol. 24 024009 (2022) – 10.1088/2040-8986/ac47b3
  • Ceron, S., Kimmel, M. A., Nilles, A. & Petersen, K. Soft Robotic Oscillators With Strain-Based Coordination. IEEE Robotics and Automation Letters vol. 6 7557–7563 (2021) – 10.1109/lra.2021.3100599
  • Ikeda, Y. et al. Coupled Oscillator Model of the Business Cycle with Fluctuating Goods Markets. Progress of Theoretical Physics Supplement vol. 194 111–121 (2012) – 10.1143/ptps.194.111
  • Hastings, A. et al. Transient phenomena in ecology. Science vol. 361 (2018) – 10.1126/science.aat6412
  • Simonsen, I., Buzna, L., Peters, K., Bornholdt, S. & Helbing, D. Transient Dynamics Increasing Network Vulnerability to Cascading Failures. Physical Review Letters vol. 100 (2008) – 10.1103/physrevlett.100.218701
  • Fan, H. et al. Learning the dynamics of coupled oscillators from transients. Physical Review Research vol. 4 (2022) – 10.1103/physrevresearch.4.013137
  • Hornik, K. Approximation capabilities of multilayer feedforward networks. Neural Networks vol. 4 251–257 (1991) – 10.1016/0893-6080(91)90009-t
  • JY Sam Greydanus. Sam Greydanus JY, Dzamba M, Yosinski J (2019) Hamiltonian neural networks. Adv Neural Inform Proc Syst 32:1110–1118 (2019)
  • Desai, S. A., Mattheakis, M., Sondak, D., Protopapas, P. & Roberts, S. J. Port-Hamiltonian neural networks for learning explicit time-dependent dynamical systems. Physical Review E vol. 104 (2021) – 10.1103/physreve.104.034312
  • Choudhary, A. et al. Forecasting Hamiltonian dynamics without canonical coordinates. Nonlinear Dynamics vol. 103 1553–1562 (2021) – 10.1007/s11071-020-06185-2
  • Li, S. & Yang, Y. A recurrent neural network framework with an adaptive training strategy for long-time predictive modeling of nonlinear dynamical systems. Journal of Sound and Vibration vol. 506 116167 (2021) – 10.1016/j.jsv.2021.116167
  • Ölmez, M. & Güzeliş, C. Exploiting Chaos in Learning System Identification for Nonlinear State Space Models. Neural Processing Letters vol. 41 29–41 (2013) – 10.1007/s11063-013-9332-7
  • L Ziyin. Ziyin L, Hartwig T, Ueda M (2020) Neural networks fail to learn periodic functions and how to fix it. Adv Neural Inform Proc Syst 33:1583–1594 (2020)
  • Mattheakis, M., Sondak, D., Dogra, A. S. & Protopapas, P. Hamiltonian neural networks for solving equations of motion. Physical Review E vol. 105 (2022) – 10.1103/physreve.105.065305
  • Tomasiello, S., Loia, V. & Khaliq, A. A granular recurrent neural network for multiple time series prediction. Neural Computing and Applications vol. 33 10293–10310 (2021) – 10.1007/s00521-021-05791-4
  • Colace, F., Loia, V. & Tomasiello, S. Revising recurrent neural networks from a granular perspective. Applied Soft Computing vol. 82 105535 (2019) – 10.1016/j.asoc.2019.105535
  • Choudhary, A. et al. Physics-enhanced neural networks learn order and chaos. Physical Review E vol. 101 (2020) – 10.1103/physreve.101.062207
  • Han, C.-D., Glaz, B., Haile, M. & Lai, Y.-C. Adaptable Hamiltonian neural networks. Physical Review Research vol. 3 (2021) – 10.1103/physrevresearch.3.023156
  • Zhang, H., Fan, H., Wang, L. & Wang, X. Learning Hamiltonian dynamics with reservoir computing. Physical Review E vol. 104 (2021) – 10.1103/physreve.104.024205
  • Desai, S. A., Mattheakis, M. & Roberts, S. J. Variational integrator graph networks for learning energy-conserving dynamical systems. Physical Review E vol. 104 (2021) – 10.1103/physreve.104.035310
  • Shougat, Md. R. E. U., Li, X., Mollik, T. & Perkins, E. An Information Theoretic Study of a Duffing Oscillator Array Reservoir Computer. Journal of Computational and Nonlinear Dynamics vol. 16 (2021) – 10.1115/1.4051270
  • H Goldstein. Goldstein H, Poole C, Safko J (2002) Classical mechanics, 3rd edn. Pearson Addison Wesley, USA (2002)
  • Rajasekar, S. & Raj, S. P. The Painlevé property, integrability and chaotic behaviour of a two-coupled Duffing oscillators. Pramana vol. 47 183–198 (1996) – 10.1007/bf02847763
  • Hong, Y. P. & Pan, C.-T. A lower bound for the smallest singular value. Linear Algebra and its Applications vol. 172 27–32 (1992) – 10.1016/0024-3795(92)90016-4
  • Tomasiello, S., Pedrycz, W. & Loia, V. Contemporary Fuzzy Logic. Big and Integrated Artificial Intelligence (Springer International Publishing, 2022). doi:10.1007/978-3-030-98974-3 – 10.1007/978-3-030-98974-3
  • Virtanen, P. et al. SciPy 1.0: fundamental algorithms for scientific computing in Python. Nature Methods vol. 17 261–272 (2020) – 10.1038/s41592-019-0686-2
  • Loia, V., Tomasiello, S., Vaccaro, A. & Gao, J. Using local learning with fuzzy transform: application to short term forecasting problems. Fuzzy Optimization and Decision Making vol. 19 13–32 (2019) – 10.1007/s10700-019-09311-x
  • Hastings, A. Transient dynamics and persistence of ecological systems. Ecology Letters vol. 4 215–220 (2001) – 10.1046/j.1461-0248.2001.00220.x
  • Courtney, S. M., Ungerleider, L. G., Keil, K. & Haxby, J. V. Transient and sustained activity in a distributed neural system for human working memory. Nature vol. 386 608–611 (1997) – 10.1038/386608a0