×

zbMATH — the first resource for mathematics

Symplectic Gaussian process regression of maps in Hamiltonian systems. (English) Zbl 1471.37072
Summary: We present an approach to construct structure-preserving emulators for Hamiltonian flow maps and Poincaré maps based directly on orbit data. Intended applications are in moderate-dimensional systems, in particular, long-term tracing of fast charged particles in accelerators and magnetic plasma confinement configurations. The method is based on multi-output Gaussian process (GP) regression on scattered training data. To obtain long-term stability, the symplectic property is enforced via the choice of the matrix-valued covariance function. Based on earlier work on spline interpolation, we observe derivatives of the generating function of a canonical transformation. A product kernel produces an accurate implicit method, whereas a sum kernel results in a fast explicit method from this approach. Both are related to symplectic Euler methods in terms of numerical integration but fulfill a complementary purpose. The developed methods are first tested on the pendulum and the Hénon-Heiles system and results compared to spectral regression of the flow map with orthogonal polynomials. Chaotic behavior is studied on the standard map. Finally, the application to magnetic field line tracing in a perturbed tokamak configuration is demonstrated. As an additional feature, in the limit of small mapping times, the Hamiltonian function can be identified with a part of the generating function and thereby learned from observed time-series data of the system’s evolution. For implicit GP methods, we demonstrate regression performance comparable to spectral bases and artificial neural networks for symplectic flow maps, applicability to Poincaré maps, and correct representation of chaotic diffusion as well as a substantial increase in performance for learning the Hamiltonian function compared to existing approaches.
©2021 American Institute of Physics
MSC:
37M15 Discretization methods and integrators (symplectic, variational, geometric, etc.) for dynamical systems
65P10 Numerical methods for Hamiltonian systems including symplectic integrators
PDF BibTeX XML Cite
Full Text: DOI
References:
[1] Goldstein, H., Classical Mechanics (1980), Addison-Wesley · Zbl 0491.70001
[2] Arnold, V. I., Mathematical Methods of Classical Mechanics, Graduate Texts in Mathematics Vol. 60 (Springer, New York, 1989).
[3] Marsden, J. E.; Ratiu, T. S., Introduction to Mechanics and Symmetry (1999), Springer: Springer, New York
[4] Neil, R. M., “MCMC using Hamiltonian dynamics,” in Handbook of Markov Chain Monte Carlo, edited by S. Brooks, A. Gelman, G. L. Jones, and X.-L. Meng (Chapman & Hall/CRC, 2011), pp. 113-162.
[5] Lichtenberg, A. and Lieberman, M., Regular and Chaotic Dynamics, Applied Mathematical Sciences (Springer, 1992). · Zbl 0748.70001
[6] Hairer, E.; Lubich, C.; Wanner, G., Geometric Numerical Integration: Structure-Preserving Algorithms for Ordinary Differential Equations (2006), Springer · Zbl 1094.65125
[7] Casas, F.; Blanes, S., A Concise Introduction to Geometric Numerical Integration (2016), Chapman and Hall/CRC · Zbl 1343.65146
[8] McLachlan, R. I.; Quispel, G. R. W., Geometric integrators for ODEs, J. Phys. A: Math. Gen., 39, 5251-5285 (2006) · Zbl 1092.65055
[9] McLachlan, R. I.; Quispel, G. R. W., Splitting methods, Acta Numer., 11, 341-434 (2002) · Zbl 1105.65341
[10] Quispel, G. R. W.; McLaren, D. I., A new class of energy-preserving numerical integration methods, J. Phys. A: Math. Theor., 41, 045206 (2008) · Zbl 1132.65065
[11] Danieli, C.; Manda, B. M.; Thudiyangal, M.; Skokos, C.
[12] Abdullaev, S. S., Construction of Mappings for Hamiltonian Systems and Their Applications (2006), Springer · Zbl 1106.70001
[13] Berg, J. S.; Warnock, R. L.; Ruth, R. D.; Forest, É., Construction of symplectic maps for nonlinear motion of particles in accelerators, Phys. Rev. E, 49, 722-739 (1994)
[14] Kasilov, S. V.; Moiseenko, V. E.; Heyn, M. F., Solution of the drift kinetic equation in the regime of weak collisions by stochastic mapping techniques, Phys. Plasmas, 4, 2422 (1997)
[15] Kasilov, S. V.; Kernbichler, W.; Nemov, V. V.; Heyn, M. F., Mapping technique for stellarators, Phys. Plasmas, 9, 3508 (2002)
[16] Warnock, R., Cai, Y., and Ellison, J. A., “Construction of large period symplectic maps by interpolative methods,” Report No. SLAC-PUB-13867, SLAC National Accelerator Laboratory, 2009.
[17] Bertalan, T.; Dietrich, F.; Mezić, I.; Kevrekidis, I. G., On learning Hamiltonian systems from data, Chaos, 29, 121107 (2019)
[18] Burby, J. W.; Tang, Q.; Maulik, R.
[19] Deco, G.; Brauer, W., Nonlinear higher-order statistical decorrelation by volume-conserving neural architectures, Neural Netw., 8, 525-535 (1995)
[20] Parra, L. C., “Symplectic nonlinear component analysis,” in Proceedings of the 8th International Conference on Neural Information Processing Systems: NIPS’95 (MIT Press, Cambridge, MA, 1995), pp. 437-443.
[21] Greydanus, S.; Dzamba, M.; Yosinski, J.
[22] Chen, Z.; Zhang, J.; Arjovsky, M.; Bottou, L.
[23] Toth, P.; Rezende, D. J.; Jaegle, A.; Racanière, S.; Botev, A.; Higgins, I.
[24] Jin, P.; Zhang, Z.; Zhu, A.; Tang, Y.; Karniadakis, G. E.
[25] Parra, L. C.
[26] Li, S.-H.; Dong, C.-X.; Zhang, L.; Wang, L., Neural canonical transformation with symplectic flows, Phys. Rev. X, 10, 021020 (2020)
[27] Hamzi, B.; Owhadi, H.
[28] Raissi, M.; Perdikaris, P.; Karniadakis, G. E., Inferring solutions of differential equations using noisy multi-fidelity data, J. Comput. Phys., 335, 736-746 (2017) · Zbl 1382.65229
[29] Rasmussen, C. E.; Williams, C. K. I., Gaussian Processes for Machine Learning (Adaptive Computation and Machine Learning) (2005), The MIT Press
[30] Solak, E., Murray-Smith, R., Leithead, W. E., Leith, D. J., and Rasmussen, C. E., “Derivative observations in Gaussian process models of dynamic systems,” in Advances in Neural Information Processing Systems 15, edited by S. Becker, S. Thrun, and K. Obermayer (MIT Press, 2003), pp. 1057-1064.
[31] Eriksson, D.; Lee, E.; Dong, K.; Bindel, D.; Wilson, A., Scaling Gaussian process regression with derivatives, Adv. Neural Inf. Process. Syst., 2018, 6867-6877
[32] O’Hagan, A., Some Bayesian numerical analysis, Bayesian Stat., 4, 345-363 (1992)
[33] Álvarez, M. A.; Rosasco, L.; Lawrence, N. D., Kernels for vector-valued functions: A review, Found. Trends Mach. Learn., 4, 195-266 (2012) · Zbl 1301.68212
[34] Boozer, A. H., Physics of magnetically confined plasmas, Rev. Mod. Phys., 76, 1071-1141 (2005)
[35] Littlejohn, R. G., Variational principles of guiding centre motion, J. Plasma Phys., 29, 111-125 (1983)
[36] Cary, J. R.; Brizard, A. J., Hamiltonian theory of guiding-center motion, Rev. Mod. Phys., 81, 693-738 (2009) · Zbl 1205.37068
[37] White, R., The Theory of Toroidally Confined Plasmas (2006), Imperial College Press · Zbl 1213.76003
[38] Cary, J. R.; Littlejohn, R. G., Noncanonical Hamiltonian mechanics and its application to magnetic field line flow, Ann. Phys., 151, 1-34 (1983)
[39] Boozer, A. H., Time-dependent drift Hamiltonian, Phys. Fluids, 27, 2441-2445 (1984) · Zbl 0546.76153
[40] Li, M.; Breizman, B. N.; Zheng, L., Canonical straight field line magnetic flux coordinates for tokamaks, J. Comput. Phys., 326, 334-341 (2016)
[41] Albert, C.; Kasilov, S.; Kernbichler, W., Symplectic integration with non-canonical quadrature for guiding-center orbits in magnetic confinement devices, J. Comput. Phys., 403, 109065 (2020) · Zbl 1453.65436
[42] Albert, C. G.; Rath, K., Gaussian process regression for data fulfilling linear differential equations with localized sources, Entropy, 22, 152 (2020)
[43] Fasshauer, G. E., “Solving partial differential equations by collocation with radial basis functions,” in Surface Fitting and Multiresolution Methods, edited by A. L. Méhauté, C. Rabut, and L. Schumaker (Vanderbilt University Press, 1997), Vol. 2, pp. 131-138. · Zbl 0938.65140
[44] Owhadi, H., Bayesian numerical homogenization, Multiscale Model. Simul., 13, 812-828 (2015) · Zbl 1322.35002
[45] Park, J.; Sandberg, I. W., Universal approximation using radial-basis-function networks, Neural Comput., 3, 246-257 (1991)
[46] Micchelli, C. A.; Xu, Y.; Zhang, H., Universal kernels, J. Mach. Learn. Res., 7, 2651-2667 (2006) · Zbl 1222.68266
[47] Caponnetto, A.; Micchelli, C. A.; Pontil, M.; Ying, Y., Universal multi-task kernels, J. Mach. Learn. Res., 9, 1615-1646 (2008) · Zbl 1225.68155
[48] Zhu, C.; Byrd, R. H.; Lu, P.; Nocedal, J., Algorithm 778: L-bfgs-b: Fortran subroutines for large-scale bound-constrained optimization, ACM Trans. Math. Softw., 23, 550-560 (1997) · Zbl 0912.65057
[49] Hansen, N.
[50] Hansen, N., Akimoto, Y., and Baudis, P., CMA-ES/pycma on GitHub, Zenodo (2019).
[51] Niederreiter, H., Random Number Generation and Quasi-Monte Carlo Methods (1992), Society for Industrial and Applied Mathematics: Society for Industrial and Applied Mathematics, Philadelphia, PA · Zbl 0761.65002
[52] Dormand, J.; Prince, P., A family of embedded Runge-Kutta formulae, J. Comput. Appl. Math., 6, 19-26 (1980) · Zbl 0448.65045
[53] Hénon, M.; Heiles, C., The applicability of the third integral of motion: Some numerical experiments, Astron. J., 69, 73-79 (1964)
[54] Zotos, E. E., An overview of the escape dynamics in the Hénon-Heiles Hamiltonian system, Meccanica, 52, 2615-2630 (2017) · Zbl 1380.37138
[55] Chirikov, B. V., A universal instability of many-dimensional oscillator systems, Phys. Rep., 52, 263-379 (1979)
[56] Rechester, A. B.; White, R. B., Calculation of turbulent diffusion for the Chirikov-Taylor model, Phys. Rev. Lett., 44, 1586-1589 (1980) · Zbl 1404.82059
[57] Ichikawa, Y. H.; Kamimura, T.; Hatori, T., Stochastic diffusion in the standard map, Physica D, 29, 247-255 (1987)
[58] Dana, I.; Fishman, S., Diffusion in the standard map, Physica D, 17, 63-74 (1985) · Zbl 0595.58015
[59] Venegeroles, R., Calculation of superdiffusion for the Chirikov-Taylor model, Phys. Rev. Lett., 101, 054102 (2008)
[60] Zaslavsky, G. M.; Edelman, M.; Niyazov, B. A., Self-similarity, renormalization, and phase space nonuniformity of Hamiltonian chaotic dynamics, Chaos, 7, 159-181 (1997) · Zbl 0933.37024
[61] Harsoula, M.; Contopoulos, G., Global and local diffusion in the standard map, Phys. Rev. E, 97, 022215 (2018)
[62] Manos, T.; Robnik, M., Dynamical localization in chaotic systems: Spectral statistics and localization measure in the kicked rotator as a paradigm for time-dependent and time-independent systems, Phys. Rev. E, 87, 062905 (2013)
[63] Manos, T.; Robnik, M., Survey on the role of accelerator modes for anomalous diffusion: The case of the standard map, Phys. Rev. E, 89, 022905 (2014)
[64] Eder, M.; Albert, C. G.; Bauer, L. M. P.; Kasilov, S. V.; Kernbichler, W., Quasi-geometric integration of guiding-center orbits in piecewise linear toroidal fields, Phys. Plasmas, 27, 122508 (2020)
[65] Rath, K.; Albert, C.; Bischl, B.; von Toussaint, U., Zenodo (2021)
[66] Albert, C.; Hofmeister, R.; Rath, K., Zenodo (2020)
[67] Burby, J. W.; Ellison, C. L., Toroidal regularization of the guiding center Lagrangian, Phys. Plasmas, 24, 110703 (2017)
[68] Morrison, P. J., Structure and structure-preserving algorithms for plasma physics, Phys. Plasmas, 24, 055502 (2017)
[69] Berndt, R., Klucznik, M., and Society, A. M., An Introduction to Symplectic Geometry, Contemporary Mathematics (American Mathematical Society, 2001).
[70] José, J. V.; Saletan, E. J., Classical Dynamics: A Contemporary Approach (1998), Cambridge University Press · Zbl 0918.70001
[71] Seber, G. and Lee, A., Linear Regression Analysis, Wiley Series in Probability and Statistics (Wiley, 2012). · Zbl 1029.62059
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.