Abstract
Transfer operators such as the Perron–Frobenius or Koopman operator play an important role in the global analysis of complex dynamical systems. The eigenfunctions of these operators can be used to detect metastable sets, to project the dynamics onto the dominant slow processes, or to separate superimposed signals. We propose kernel transfer operators, which extend transfer operator theory to reproducing kernel Hilbert spaces and show that these operators are related to Hilbert space representations of conditional distributions, known as conditional mean embeddings. The proposed numerical methods to compute empirical estimates of these kernel transfer operators subsume existing data-driven methods for the approximation of transfer operators such as extended dynamic mode decomposition and its variants. One main benefit of the presented kernel-based approaches is that they can be applied to any domain where a similarity measure given by a kernel is available. Furthermore, we provide elementary results on eigendecompositions of finite-rank RKHS operators. We illustrate the results with the aid of guiding examples and highlight potential applications in molecular dynamics as well as video and text data analysis.
Similar content being viewed by others
Notes
Although this term is commonly used in the literature, \( \Phi \) is technically not a matrix, but a row vector in \( \mathbb {H}^n \). If \(\mathbb {H}\) is finite dimensional, \( \Phi \) can be viewed as a matrix.
For specific choices of \(\mathbb {P}(X)\) and \(\mathbb {P}(Y)\), the boundedness assumption can be replaced by a more general integrability assumption, i.e., \(\mathbb {E}_{\scriptscriptstyle X}[k(X,X)] < \infty \) and \(\mathbb {E}_{\scriptscriptstyle Y}[l(Y,Y)] < \infty \), so that \(\mathbb {H} \subset L^2({\mathbb {X}}, \mathbb {P}(X))\) and \(\mathbb {\mathbb {G}} \subset L^2({\mathbb {Y}}, \mathbb {P}(Y))\), respectively.
An observable could be, for example, a measurement or sensor probe.
This holds, for instance, for finite domains equipped with characteristic kernels, but not necessarily for continuous domains (Song et al. 2013).
See, e.g., Kloeden and Platen (2011).
Parts of the same articles are used several times to increase the size of the data set, this is thus a synthetic example, mainly to illustrate the concept.
We use the String Kernel Software implementation.
References
Bach, F.R., Jordan, M.I.: Kernel independent component analysis. J. Mach. Learn. Res. 3, 1–48 (2003)
Baker, C.: Mutual information for Gaussian processes. SIAM J. Appl. Math. 19(2), 451–458 (1970)
Baker, C.: Joint measures and cross-covariance operators. Trans. Am. Math. Soc. 186, 273–289 (1973)
Baxter, J.R., Rosenthal, J.S.: Rates of convergence for everywhere-positive Markov chains. Stat. Probab. Lett. 22(4), 333–338 (1995)
Berlinet, A., Thomas-Agnan, C.: Reproducing Kernel Hilbert Spaces in Probability and Statistics. Kluwer Academic Publishers, Berlin (2004)
Berry, T., Giannakis, D., Harlim, J.: Nonparametric forecasting of low-dimensional dynamical systems. Phys. Rev. E 91, 032915 (2015). https://doi.org/10.1103/PhysRevE.91.032915
Bittracher, A., Koltai, P., Klus, S., Banisch, R., Dellnitz, M., Schütte, C.: Transition manifolds of complex metastable systems: theory and data-driven computation of effective dynamics. Accepted for publication in JNLS, (2017)
Bo, L., Ren, X., Fox, D.: Kernel descriptors for visual recognition. In J. D. Lafferty, C. K. I. Williams, J. Shawe-Taylor, R. S. Zemel, and A. Culotta, editors, Advances in Neural Information Processing Systems 23, pages 244–252. Curran Associates, Inc., (2010)
Brunton, S.L., Brunton, B.W., Proctor, J.L., Kutz, J.N.: Koopman invariant subspaces and finite linear representations of nonlinear dynamical systems for control. PLoS ONE 11(2), 1–19 (2016)
Case, D.A., Berryman, J.T., Betz, R.M., Cerutti, D.S., Cheatham, T.E., Darden, T.A., Duke, R.E., Giese, T.J., Gohlke, H., Goetz, A.W., Homeyer, N., Izadi, S., Janowski, P., Kaus, J., Kovalenko, A., Lee, T.S., LeGrand, S., Li, P., Luchko, T., Luo, R., Madej, B., Merz, K.M., Monard, G., Needham, P., Nguyen, H., Nguyen, H.T., Omelyan, I., Onufriev, A., Roe, D.R., Roitberg, A., Salomon-Ferrer, R., Simmerling, C.L., Smith, W., Swails, J., Walker, R.C., Wang, J., Wolf, R.M., Wu, X., York, D.M., Kollman, P.A.: AMBER 2015. University of California, San Francisco (2015)
Diestel, J., Uhl, J.J.: Vector Measures. American Mathematical Society, Providence (1977)
Dinculeanu, N.: Vector Integration and Stochastic Integration in Banach Spaces. Wiley, Hoboken (2000)
Fukumizu, K., Bach, F.R., Jordan, M.I.: Dimensionality reduction for supervised learning with Reproducing Kernel Hilbert Spaces. J. Mach. Learn. Res. 5, 73–99 (2004)
Fukumizu, K., Bach, F., Gretton, A.: Statistical consistency of kernel canonical correlation analysis. J. Mach. Learn. Res. 8, 361–383 (2007a)
Fukumizu, K., Song, L., Gretton, A.: Kernel Bayes’ rule: Bayesian inference with positive definite kernels. J. Mach. Learn. Res. 14, 3753–3783 (2013)
Giannakis, D., Das, S., Slawinska, J.: Reproducing kernel Hilbert space compactification of unitary evolution groups. arXiv e-prints, (2018)
Grünewälder, S., Lever, G., Gretton, A., Baldassarre, L., Patterson, S., Pontil, M.: Conditional mean embeddings as regressors. In: Proceedings of the 29th International Conference on Machine Learning (ICML), (2012)
Hofmann, T., Schölkopf, B., Smola, A.: Kernel methods in machine learning. Ann. Stat. 36(3), 1171–1220 (2008)
Huisinga, W.: Metastability of Markovian systems: a transfer operator based approach in application to molecular dynamics. PhD thesis, Freie Universität Berlin (2001)
Kato, T.: Perturbation Theory for Linear Operators. Springer, Berlin (1980)
Kawahara, Y.: Dynamic mode decomposition with reproducing kernels for Koopman spectral analysis. In: Advances in Neural Information Processing Systems 29, pp. 911–919. Curran Associates, Inc., (2016)
Kloeden, P.E., Platen, E.: Numerical Solution of Stochastic Differential Equations. Stochastic Modelling and Applied Probability. Springer, Berlin (2011)
Klus, S., Koltai, P., Schütte, C.: On the numerical approximation of the Perron–Frobenius and Koopman operator. J. Comput. Dyn. 3(1), 51–79 (2016)
Klus, S., Nüske, F., Koltai, P., Wu, H., Kevrekidis, I., Schütte, C., Noé, F.: Data-driven model reduction and transfer operator approximation. J. Nonlinear Sci. 28, 985–1010 (2018a). https://doi.org/10.1007/s00332-017-9437-7
Klus, S., Bittracher, A., Schuster, I., Schütte, C.: A kernel-based approach to molecular conformation analysis. J. Chem. Phys. 149, 244109 (2018b)
Klus, S., Peitz, S., Schuster, I.: Analyzing high-dimensional time-series data using kernel transfer operator eigenfunctions. ArXiv e-prints, (2018c)
Korda, M., Mezić, I.: On convergence of extended dynamic mode decomposition to the Koopman operator. J. Nonlinear Sci. 28(2), 687–710 (2018)
Lasota, A., Mackey, M. C.: Chaos, fractals, and noise: Stochastic aspects of dynamics, volume 97 of Applied Mathematical Sciences. Springer, 2nd edition, (1994)
Lodhi, H., Saunders, C., Shawe-Taylor, J., Cristianini, N., Watkins, C.: Text classification using string kernels. J. Mach. Learn. Res. 2, 419–444 (2002)
McGibbon, R.T., Pande, V.S.: Variational cross-validation of slow dynamical modes in molecular kinetics. J. Chem. Phy. 142(12), 124105 (2015)
Molgedey, L., Schuster, H.G.: Separation of a mixture of independent signals using time delayed correlations. Phys. Rev. Lett. 72, 3634–3637 (1994)
Muandet, K., Fukumizu, K., Sriperumbudur, B., Schölkopf, B.: Kernel mean embedding of distributions: a review and beyond. Found. Trends Mach. Learn. 10(1–2), 1–141 (2017)
Noé, F., Nüske, F.: A variational approach to modeling slow processes in stochastic dynamical systems. Multiscale Model. Simul. 11(2), 635–655 (2013)
Nüske, F., Keller, B.G., Pérez-Hernández, G., Mey, A.S.J.S., Noé, F.: Variational approach to molecular kinetics. J. Chem. Theory Comput. 10(4), 1739–1752 (2014)
Pérez-Hernández, G., Paul, F., Giorgino, T., De Fabritiis, G., Noé, F.: Identification of slow molecular order parameters for Markov model construction. J. Chem. Phys. 139(1), 015102 (2013)
Rowley, C.W., Mezić, I., Bagheri, S., Schlatter, P., Henningson, D.S.: Spectral analysis of nonlinear flows. J. Fluid Mech. 641, 115–127 (2009)
Schmid, P.J.: Dynamic mode decomposition of numerical and experimental data. J. Fluid Mech. 656, 5–28 (2010)
Schölkopf, B., Smola, A.J.: Learning with Kernels: Support Vector Machines, Regularization, Optimization and Beyond. MIT Press, Cambridge (2001)
Schölkopf, B., Smola, A., Müller, K.-R.: Nonlinear component analysis as a kernel eigenvalue problem. Neural Comput. 10(5), 1299–1319 (1998)
Schuster, I., Mollenhauer, M., Klus, S., Muandet, K.: Kernel conditional density operators. arXiv e-prints, (2019)
Schütte, C.: Conformational dynamics: modelling, theory, algorithm, and application to biomolecules. Habilitation Thesis (1999)
Schütte, C., Sarich, M.: Metastability and Markov State Models in Molecular Dynamics: Modeling, Analysis, Algorithmic Approaches. Number 24 in Courant Lecture Notes. American Mathematical Society (2013)
Schwantes, C.R., Pande, V.S.: Modeling molecular kinetics with tICA and the kernel trick. J. Chem. Theory Comput. 11(2), 600–608 (2015)
Sejdinovic, D., Strathmann, H., Garcia, M.L., Andrieu, C., Gretton, A.: Kernel adaptive Metropolis–Hastings. In: International conference on machine learning, pp. 1665–1673 (2014)
Smola, A., Gretton, A., Song, L., Schölkopf, B.: A Hilbert space embedding for distributions. In: Proceedings of the 18th International Conference on Algorithmic Learning Theory, pp. 13–31. Springer, Berlin (2007)
Song, L., Huang, J., Smola, A., Fukumizu, K.: Hilbert space embeddings of conditional distributions with applications to dynamical systems. In: Proceedings of the 26th Annual International Conference on Machine Learning, pp. 961–968 (2009)
Song, L., Fukumizu, K., Gretton, A.: Kernel embeddings of conditional distributions: a unified kernel framework for nonparametric inference in graphical models. IEEE Signal Process. Mag. 30(4), 98–111 (2013)
Sriperumbudur, B., Gretton, A., Fukumizu, K., Lanckriet, G., Schölkopf, B.: Injective Hilbert space embeddings of probability measures. In: The 21st Annual Conference on Learning Theory, pp. 111–122. Omnipress (2008)
Steinwart, I., Christmann, A.: Support Vector Machines, 1st edn. Springer Publishing Company, Incorporated, Berlin (2008)
Tu, J.H., Rowley, C.W., Luchtenburg, D.M., Brunton, S.L., Kutz, J.N.: On dynamic mode decomposition: theory and applications. J. Comput. Dyn. 1(2), 391 (2014)
Ulam, S .M.: A Collection of Mathematical Problems. Interscience Publisher, New York (1960)
Williams, M.O., Kevrekidis, I.G., Rowley, C.W.: A data-driven approximation of the Koopman operator: extending dynamic mode decomposition. J. Nonlinear Sci. 25(6), 1307–1346 (2015a)
Williams, M.O., Rowley, C.W., Kevrekidis, I.G.: A kernel-based method for data-driven Koopman spectral analysis. J. Comput. Dyn. 2(2), 247–265 (2015b)
Acknowledgements
This research has been partially funded by Deutsche Forschungsgemeinschaft (DFG) through grant CRC 1114 “Scaling Cascades in Complex Systems”. Krikamol Muandet acknowledges fundings from the Faculty of Science, Mahidol University and the Thailand Research Fund (TRF). We would like to thank the reviewers for their helpful comments.
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by Dr. Anthony Bloch.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Klus, S., Schuster, I. & Muandet, K. Eigendecompositions of Transfer Operators in Reproducing Kernel Hilbert Spaces. J Nonlinear Sci 30, 283–315 (2020). https://doi.org/10.1007/s00332-019-09574-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00332-019-09574-z
Keywords
- Reproducing kernel Hilbert spaces
- Koopman operator
- Perron–Frobenius operator
- Eigendecompositions
- Kernel mean embeddings