A Physics-Informed LSSVR Method with Legendre Kernels for Direct Solution of Fokker-Planck Equations

Document Type : Research Article

Authors

1 Department of Computer and Data Sciences, Faculty of Mathematical Sciences, Shahid Beheshti University, Tehran, Iran

2 Department of Computer and Data Sciences, Faculty of Mathematical Sciences, Shahid Beheshti University, Tehran, Iran; Institute for Cognitive and Brain Sciences, Shahid Beheshti University, Tehran, Iran

10.22128/ansne.2025.3047.1151

Abstract

The Fokker–Planck equation is widely used to describe how systems evolve when randomness plays a role. It appears in many fields, including physics, finance, biology, and engineering. Classical numerical methods usually require discretization, which can make the computation expensive, unstable, or less accurate. In this work, we present a direct method for solving these equations using Least Squares Support Vector Regression (LSSVR) with Legendre kernels. Our approach avoids discretization and provides global optimization, which helps overcome the difficulties faced by loss-based methods such as Physics-Informed Neural Networks (PINNs). The use of Legendre kernels gives strong approximation properties and ensures high accuracy in the solutions. We tested the method on several problems and found that it achieves very precise results while being faster and more stable than PINNs. To further improve reliability, we also applied automatic hyperparameter tuning, which adapts the method to each problem without manual adjustment. These results show that LSSVR with Legendre kernels is a simple, accurate, and efficient tool for scientists and engineers who need to solve Fokker–Planck equations.

Keywords

Main Subjects

Full Text

 

Article PDF

References

[1] H. Risken, Fokker-planck equation. Springer, (1989).
[2] C. Gardiner, Stochastic methods, vol. 4. Springer Berlin Heidelberg, (2009).
[3] R. Cont and P. Tankov, Financial modelling with jump processes. Chapman and Hall/CRC, (2003).
[4] J. D. Murray, Mathematical biology: I. An introduction, vol. 17. Springer Science & Business Media, (2007).
[5] W. Coffey and Y. P. Kalmykov, The Langevin equation: with applications to stochastic problems in physics, chemistry and electrical engineering, vol. 27. World Scientific, (2012).
[6] C. Soize, Uncertainty quantification, vol. 23. Springer, (2017).
[7] T. D. Frank, Nonlinear Fokker-Planck equations: fundamentals and applications. Springer, (2005).
[8] G. A. Pavliotis, Stochastic processes and applications, 60, (2014).
[9] J. Chang and G. Cooper, A practical difference scheme for fokker-planck equations, Journal of Computational Physics, 6(1), 1–16, (1970).
[10] P. Kumar and S. Narayanan, Solution of fokker-planck equation by finite element and finite difference methods for nonlinear systems, Sadhana, 31(4), 445–461, (2006).
[11] D. Gottlieb and S. A. Orszag, Numerical analysis of spectral methods: theory and applications. SIAM, (1977).
[12] E. Platen and N. Bruti-Liberati, Numerical solution of stochastic differential equations with jumps in finance, vol. 64. Springer Science & Business Media, (2010).
[13] B. Oksendal, Stochastic differential equations: an introduction with applications. Springer Science & Business Media, (2013).
[14] J. A. Suykens and J. Vandewalle, Least squares support vector machine classifiers, vol. 9. Springer, (1999).
[15] S. Mehrkanoon, S. Mehrkanoon, and J. A. Suykens, Parameter estimation of delay differential equations: an integration-free ls-svm approach, Communications in Nonlinear Science and Numerical Simulation, 19(4), 830–841, (2014).
[16] A. Shidfar, A. Molabahrami, A. Babaei, and A. Yazdanian, A series solution of the nonlinear volterra and fredholm integro-differential equations, Communications in Nonlinear Science and Numerical Simulation, 15(2), 205–215, (2010).
[17] Z. Behdani and M. Darehmiraki, Enhancing kernel ridge regression models with compact support wendland functions, Analytical and Numerical Solutions for Nonlinear Equations, 9(1), 34–47, (2025).
[18] J. A. Rad, K. Parand, and S. Chakraverty, Learning with fractional orthogonal kernel classifiers in support vector machines: Theory, algorithms and applications. Springer, (2023).
[19] R. Pourgholi, H. Dana, and S. H. Tabasi, Solving an inverse heat conduction problem using genetic algorithm: sequential and multi-core parallelization approach, Applied Mathematical Modelling, 38(7-8), 1948–1958, (2014).
[20] H. Dana Mazraeh, M. Kalantari, S. H. Tabasi, A. Afzal Aghaei, Z. Kalantari, and F. Fahimi, Solving fredholm integral equations of the second kind using an improved cuckoo optimization algorithm, Analytical and Numerical Solutions for Nonlinear Equations, 7(1), 33–52, (2022).
[21] H. Dana Mazraeh, K. Parand, H. Farahani, and S. Kheradpisheh, An improved imperialist competitive algorithm for solving an inverse form of the huxley equation, Iranian Journal of Numerical Analysis and Optimization, 14(3), 681–707, (2024).
[22] E. Mirzabeigi, S. Rezaee, and K. Parand, Lyam: Robust non-convex optimization for stable learning in noisy environments, arXiv preprint arXiv:2507.11262, (2025).
[23] M. Raissi, P. Perdikaris, and G. E. Karniadakis, Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations, 378, 686–707, (2019).
[24] K. M. Mohammadi, M. Babaei, Z. Hajimohammadi, and K. Parand, A new numerical method for solving neuro-cognitive models via chebyshev deep neural network (cdnn), Applications of Mathematics, 1–19, (2025).
[25] S. Wang, Y. Teng, and P. Perdikaris, Understanding and mitigating gradient flow pathologies in physics-informed neural networks, SIAM Journal on Scientific Computing, 43(5), A3055–A3081, (2021).
[26] E. Mirzabeigi, R. Salehi, and K. Parand, Bridgenet: A hybrid, physics-informed machine learning framework for solving high-dimensional fokker-planck equations, arXiv preprint arXiv:2506.04354, (2025).
[27] E. Mirzabeigi, M. Babaei, A. H. Karami, S. Rezaee, R. Salehi, and K. Parand, Physics-informed lane-emden solvers using lynx-net: Implementing radial basis functions in kolmogorov representation, Astronomy and Computing, 100997, (2025).
[28] J. P. Boyd, Chebyshev and Fourier spectral methods. Courier Corporation, (2001).
[29] C. Canuto, M. Y. Hussaini, A. Quarteroni, and T. A. Zang, Spectral methods: fundamentals in single domains. Springer, (2006).
[30] J. Shen, T. Tang, and L.-L. Wang, Spectral methods: algorithms, analysis and applications, vol. 41. Springer Science & Business Media, (2011).
[31] S. Mehrkanoon and J. A. Suykens, Learning solutions to partial differential equations using ls-svm, 159, 105–116, (2015).
[32] M. Razzaghi and S. Yousefi, The legendre wavelets operational matrix of integration, International journal of systems science, 32(4), 495–502, (2001).
[33] M. Babaei, A. A. Aghaei, Z. Kazemi, M. Jamshidi, R. Ghaderi, and K. Parand, Solving a class of thomas–fermi equations: A new solution concept based on physics-informed machine learning, 225, 716–730, (2024).
[34] S. Mirzaei and A. Shokri, A pseudo-spectral approach to solving the fractional cable equation using lagrange polynomials, Analytical and Numerical Solutions for Nonlinear Equations, 9(1), 20–33, (2025).
[35] R. Burden, J. Faires, and A. Burden, Numerical Analysis. Cengage Learning, (2015).
[36] P. Ahadian and K. Parand, Support vector regression for the temperature-stimulated drug release, 165, 112871, (2022).
[37] T. Akiba, S. Sano, T. Yanase, T. Ohta, and M. Koyama, Optuna: A next-generation hyperparameter optimization framework, in Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery & data mining, 2623–2631, (2019).
[38] M. Lakestani and M. Dehghan, Numerical solution of fokker-planck equation using the cubic b-spline scaling functions, Numerical Methods for Partial Differential Equations: An International Journal, 25(2), 418–429, (2009).
[39] H. Dana Mazraeh, P. Motaharinezhad, N. Daneshian, and K. Parand, Solving the fokker-planck equation with neural networks: A performance improvement approach, Analytical and Numerical Solutions for Nonlinear Equations, 9(1), 1–11, (2025).
[40] M. Tatari, M. Dehghan, and M. Razzaghi, Application of the adomian decomposition method for the fokker–planck equation, Mathematical and Computer Modelling, 45(5-6), 639–650, (2007).
[41] S. Kazem, J. Rad, and K. Parand, Radial basis functions methods for solving fokker–planck equation, Engineering Analysis with Boundary Elements, 36(2), 181–189, (2012). 
Analytical and Numerical Solutions for Nonlinear Equations
Volume 9, Issue 2
September 2025
Pages 193-202

Files

History

  • Receive Date: 07 September 2025
  • Revise Date: 17 September 2025
  • Accept Date: 22 September 2025
  • Publish Date: 27 September 2025

Share

How to cite

Statistics