Papers
Topics
Authors
Recent
Assistant
AI Research Assistant
Well-researched responses based on relevant abstracts and paper content.
Custom Instructions Pro
Preferences or requirements that you'd like Emergent Mind to consider when generating responses.
Gemini 2.5 Flash
Gemini 2.5 Flash 134 tok/s
Gemini 2.5 Pro 46 tok/s Pro
GPT-5 Medium 23 tok/s Pro
GPT-5 High 32 tok/s Pro
GPT-4o 101 tok/s Pro
Kimi K2 179 tok/s Pro
GPT OSS 120B 435 tok/s Pro
Claude Sonnet 4.5 36 tok/s Pro
2000 character limit reached

Resolvent-Type Data-Driven Learning of Generators for Unknown Continuous-Time Dynamical Systems (2411.00923v1)

Published 1 Nov 2024 in math.DS

Abstract: A semigroup characterization, or equivalently, a characterization by the generator, is a classical technique used to describe continuous-time nonlinear dynamical systems. In the realm of data-driven learning for an unknown nonlinear system, one must estimate the generator of the semigroup of the system's transfer operators (also known as the semigroup of Koopman operators) based on discrete-time observations and verify convergence to the true generator in an appropriate sense. As the generator encodes essential transient transitional information of the system, challenges arise for some existing methods that rely on accurately estimating the time derivatives of the state with constraints on the observation rate. Recent literature develops a technique that avoids the use of time derivatives by employing the logarithm of a Koopman operator. However, the validity of this method has been demonstrated only within a restrictive function space and requires knowledge of the operator's spectral properties. In this paper, we propose a resolvent-type method for learning the system generator to relax the requirements on the observation frequency and overcome the constraints of taking operator logarithms. We also provide numerical examples to demonstrate its effectiveness in applications of system identification and constructing Lyapunov functions.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (52)
  1. S. N. Ethier and T. G. Kurtz, Markov Processes: Characterization and Convergence. John Wiley & Sons, 2009.
  2. B. O. Koopman, “Hamiltonian systems and transformation in Hilbert space,” Proceedings of the National Academy of Sciences, vol. 17, no. 5, pp. 315–318, 1931.
  3. Gulf Professional Publishing, 1980.
  4. Springer Science & Business Media, 2008.
  5. American Mathematical Society, 2022.
  6. A. Vannelli and M. Vidyasagar, “Maximal Lyapunov functions and domains of attraction for autonomous nonlinear systems,” Automatica, vol. 21, no. 1, pp. 69–80, 1985.
  7. I. M. Mitchell, A. M. Bayen, and C. J. Tomlin, “A time-dependent Hamilton-Jacobi formulation of reachable sets for continuous dynamic games,” IEEE Transactions on Automatic Control, vol. 50, no. 7, pp. 947–957, 2005.
  8. Springer, 1997.
  9. M. G. Crandall and P.-L. Lions, “Viscosity solutions of Hamilton-Jacobi equations,” Transactions of the American mathematical society, vol. 277, no. 1, pp. 1–42, 1983.
  10. J. Sjöberg, Q. Zhang, L. Ljung, A. Benveniste, B. Delyon, P.-Y. Glorennec, H. Hjalmarsson, and A. Juditsky, “Nonlinear black-box modeling in system identification: a unified overview,” Automatica, vol. 31, no. 12, pp. 1691–1724, 1995.
  11. R. Haber and H. Unbehauen, “Structure identification of nonlinear dynamic systems—a survey on input/output approaches,” Automatica, vol. 26, no. 4, pp. 651–677, 1990.
  12. Y. Lin, E. D. Sontag, and Y. Wang, “A smooth converse Lyapunov theorem for robust stability,” SIAM Journal on Control and Optimization, vol. 34, no. 1, pp. 124–160, 1996.
  13. A. R. Teel and L. Praly, “A smooth Lyapunov function from a class-estimate involving two positive semidefinite functions,” ESAIM: Control, Optimisation and Calculus of Variations, vol. 5, pp. 313–367, 2000.
  14. Y. Meng, Y. Li, M. Fitzsimmons, and J. Liu, “Smooth converse Lyapunov-barrier theorems for asymptotic stability with safety constraints and reach-avoid-stay specifications,” Automatica, vol. 144, p. 110478, 2022.
  15. A. D. Ames, X. Xu, J. W. Grizzle, and P. Tabuada, “Control barrier function based quadratic programs for safety critical systems,” IEEE Transactions on Automatic Control, vol. 62, no. 8, pp. 3861–3876, 2016.
  16. A. D. Ames, S. Coogan, M. Egerstedt, G. Notomista, K. Sreenath, and P. Tabuada, “Control barrier functions: Theory and applications,” in 2019 18th European Control Conference (ECC), pp. 3420–3431, IEEE, 2019.
  17. Y. Meng and J. Liu, “Lyapunov-barrier characterization of robust reach–avoid–stay specifications for hybrid systems,” Nonlinear Analysis: Hybrid Systems, vol. 49, p. 101340, 2023.
  18. I. Mezić, “Spectral properties of dynamical systems, model reduction and decompositions,” Nonlinear Dynamics, vol. 41, pp. 309–325, 2005.
  19. W. Pan, Y. Yuan, J. Gonçalves, and G.-B. Stan, “A sparse Bayesian approach to the identification of nonlinear state-space systems,” IEEE Transactions on Automatic Control, vol. 61, no. 1, pp. 182–187, 2015.
  20. S. L. Brunton, J. L. Proctor, and J. N. Kutz, “Discovering governing equations from data by sparse identification of nonlinear dynamical systems,” Proceedings of the National Academy of Sciences, vol. 113, no. 15, pp. 3932–3937, 2016.
  21. J. C. Nash and M. Walker-Smith, “Nonlinear parameter estimation,” An integrated system on BASIC. NY, Basel, vol. 493, 1987.
  22. J. M. Varah, “A spline least squares method for numerical parameter estimation in differential equations,” SIAM Journal on Scientific and Statistical Computing, vol. 3, no. 1, pp. 28–46, 1982.
  23. S. E. Otto and C. W. Rowley, “Koopman operators for estimation and control of dynamical systems,” Annual Review of Control, Robotics, and Autonomous Systems, vol. 4, no. 1, pp. 59–87, 2021.
  24. A. Mauroy and I. Mezić, “A spectral operator-theoretic framework for global stability,” in 52nd IEEE Conference on Decision and Control, pp. 5234–5239, IEEE, 2013.
  25. A. Mauroy and I. Mezić, “Global stability analysis using the eigenfunctions of the Koopman operator,” IEEE Transactions on Automatic Control, vol. 61, no. 11, pp. 3356–3369, 2016.
  26. S. A. Deka, A. M. Valle, and C. J. Tomlin, “Koopman-based neural Lyapunov functions for general attractors,” in 61st Conference on Decision and Control (CDC), pp. 5123–5128, IEEE, 2022.
  27. Y. Meng, R. Zhou, and J. Liu, “Learning regions of attraction in unknown dynamical systems via Zubov-Koopman lifting: Regularities and convergence,” arXiv preprint arXiv:2311.15119, 2023.
  28. J. J. Bramburger and G. Fantuzzi, “Auxiliary functions as Koopman observables: Data-driven analysis of dynamical systems via polynomial optimization,” Journal of Nonlinear Science, vol. 34, no. 1, p. 8, 2024.
  29. S. Klus, F. Nüske, S. Peitz, J.-H. Niemann, C. Clementi, and C. Schütte, “Data-driven approximation of the Koopman generator: Model reduction, system identification, and control,” Physica D: Nonlinear Phenomena, vol. 406, p. 132416, 2020.
  30. A. Nejati, A. Lavaei, S. Soudjani, and M. Zamani, “Data-driven estimation of infinitesimal generators of stochastic systems,” IFAC-PapersOnLine, vol. 54, no. 5, pp. 277–282, 2021.
  31. C. Wang, Y. Meng, S. L. Smith, and J. Liu, “Data-driven learning of safety-critical control with stochastic control barrier functions,” in 61st Conference on Decision and Control (CDC), pp. 5309–5315, IEEE, 2022.
  32. S. Otto, S. Peitz, and C. Rowley, “Learning bilinear models of actuated koopman generators from partially observed trajectories,” SIAM Journal on Applied Dynamical Systems, vol. 23, no. 1, pp. 885–923, 2024.
  33. A. Mauroy and J. Goncalves, “Koopman-based lifting techniques for nonlinear systems identification,” IEEE Transactions on Automatic Control, vol. 65, no. 6, pp. 2550–2565, 2019.
  34. Z. Drmač, I. Mezić, and R. Mohr, “Identification of nonlinear systems using the infinitesimal generator of the Koopman semigroup—a numerical implementation of the Mauroy–Goncalves method,” Mathematics, vol. 9, no. 17, p. 2075, 2021.
  35. M. Black and D. Panagou, “Safe control design for unknown nonlinear systems with Koopman-based fixed-time identification,” IFAC-PapersOnLine, vol. 56, no. 2, pp. 11369–11376, 2023.
  36. Z. Zeng, Z. Yue, A. Mauroy, J. Gonçalves, and Y. Yuan, “A sampling theorem for exact identification of continuous-time nonlinear dynamical systems,” IEEE Transactions on Automatic Control, 2024.
  37. Z. Zeng, J. Liu, and Y. Yuan, “A generalized nyquist-shannon sampling theorem using the Koopman operator,” IEEE Transactions on Signal Processing, 2024.
  38. Y. Susuki, A. Mauroy, and I. Mezic, “Koopman resolvent: A laplace-domain analysis of nonlinear autonomous dynamical systems,” SIAM Journal on Applied Dynamical Systems, vol. 20, no. 4, pp. 2013–2036, 2021.
  39. Y. Meng, R. Zhou, M. Ornik, and J. Liu, “Koopman-based learning of infinitesimal generators without operator logarithm,” in 63rd IEEE Conference on Decision and Control (CDC), IEEE, 2024.
  40. Springer Science & Business Media, 2012.
  41. J. C. Willems, “Dissipative dynamical systems part I: General theory,” Archive for Rational Mechanics and Analysis, vol. 45, no. 5, pp. 321–351, 1972.
  42. M. O. Williams, I. G. Kevrekidis, and C. W. Rowley, “A data–driven approximation of the Koopman operator: Extending dynamic mode decomposition,” Journal of Nonlinear Science, vol. 25, pp. 1307–1346, 2015.
  43. Y. Meng, H. Li, M. Ornik, and X. Li, “Koopman-based data-driven techniques for adaptive cruise control system identification,” in 27th IEEE International Conference on Intelligent Transportation Systems (ITSC), IEEE, 2024.
  44. M. Axenides and E. Floratos, “Scaling properties of the Lorenz system and dissipative Nambu mechanics,” in Chaos, Information Processing And Paradoxical Games: The Legacy Of John S Nicolis, pp. 27–41, World Scientific, 2015.
  45. S. Sato, M. Sano, and Y. Sawada, “Practical methods of measuring the generalized dimension and the largest lyapunov exponent in high dimensional chaotic systems,” Progress of theoretical physics, vol. 77, no. 1, pp. 1–5, 1987.
  46. E. N. Lorenz, “Predictability: A problem partly solved,” in Seminar on Predictability, vol. 1, Reading, 1996.
  47. O. Azencot, N. B. Erichson, V. Lin, and M. Mahoney, “Forecasting sequential data using consistent koopman autoencoders,” in International Conference on Machine Learning, pp. 475–485, PMLR, 2020.
  48. T. De Ryck, S. Lanthaler, and S. Mishra, “On the approximation of functions by tanh neural networks,” Neural Networks, vol. 143, pp. 732–750, 2021.
  49. J. Liu, Y. Meng, M. Fitzsimmons, and R. Zhou, “Physics-informed neural network Lyapunov functions: PDE characterization, learning, and verification,” arXiv preprint arXiv:2312.09131, 2023.
  50. F. Camilli, L. Grüne, and F. Wirth, “A generalization of Zubov’s method to perturbed systems,” SIAM Journal on Control and Optimization, vol. 40, no. 2, pp. 496–515, 2001.
  51. J. Liu, Y. Meng, M. Fitzsimmons, and R. Zhou, “Tool LyZNet: A lightweight Python tool for learning and verifying neural Lyapunov functions and regions of attraction,” in 27th ACM International Conference on Hybrid Systems: Computation and Control, pp. 1–8, 2024.
  52. R. Zhou, M. Fitzsimmons, Y. Meng, and J. Liu, “Physics-informed extreme learning machine Lyapunov functions,” IEEE Control Systems Letters, 2024.
Citations (1)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
Dice Question Streamline Icon: https://streamlinehq.com

Open Problems

  1. Convergence of data-driven Koopman-operator logarithms to generators in general nonlinear systems
Lightbulb Streamline Icon: https://streamlinehq.com

Continue Learning

We haven't generated follow-up questions for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Generate Now
List To Do Tasks Checklist Streamline Icon: https://streamlinehq.com

Collections

Sign up for free to add this paper to one or more collections.

User Circle Single Streamline Icon: https://streamlinehq.com Sign Up
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets

This paper has been mentioned in 1 tweet and received 3 likes.

Upgrade to Pro to view all of the tweets about this paper:

Start a free 7-day Pro trial