Deep inference: A convolutional neural networks method for parameter recovery of the fractional dynamics

Document Type : Research Paper


1 Faculty of Sciences, Imam Ali University, Tehran, Iran

2 Department of Cognitive Modeling, Institute for Cognitive and Brain Sciences, Shahid Beheshti University, Tehran, Iran

3 Faculty of Engineering, Imam Ali University, Tehran Iran


Parameter recovery of dynamical systems has attracted much attention in recent years. The proposed methods for this purpose can not be used in real-time applications. Besides, little works have been done on the parameter recovery of the fractional dynamics. Therefore, in this paper, a convolutional neural network is proposed for parameter recovery of the fractional dynamics. The presented network can also estimate the uncertainty of the parameter estimation and has perfect robustness for real-time applications.


[1] D. Baleanu and A. Mendes Lopes, Handbook of Fractional Calculus with Applications, De Gruyter, 2019.
[2] V.V. Kulish and J.L. Lage, Application of fractional calculus to fluid mechanics, J. Fluids Eng. 124(3) (2002) 803–806.
[3] N. Hassan Sweilam, S. Mahyoub Al-Mekhlafi, T. Assiri, and A. Atangana, Optimal control for cancer treatment mathematical model using Atangana–Baleanu–Caputo fractional derivative, Adv. Differ. Equ. 2020(1) (2020)1–21.
[4] B. Ni and H. Gao, A deep learning approach to the inverse problem of modulus identification in elasticity, MRS Bull. 46 (2021) 19-25.
[5] M. Ridha Znaidi, G. Gupta, K. Asgari, and P. Bogdan, Identifying arguments of space-time fractional diffusion: data-driven approach, Front. Appl. Math. Stat. 6 (2020) 14.
[6] F. Liu and K. Burrage, Novel techniques in parameter estimation for fractional dynamical models arising from biological systems, Comput. Math. Appl. 62(3) (2011) 822–833.
[7] Sh. Chen, F. Liu, X. Jiang, I. Turner, and K. Burrage, Fast finite difference approximation for identifying parameters in a two-dimensional space-fractional nonlocal model with variable diffusivity coefficients, SIAM J. Numer. Anal. 54(2) (2016) 606–624.
[8] M. Raissi, P. Perdikaris, and G.E. Karniadakis, Machine learning of linear differential equations using Gaussian processes, J. Comput. Phys. 348 (2017) 683–693.
[9] M. Raissi, P. Perdikaris, and G.E. Karniadakis, Numerical Gaussian processes for time-dependent and nonlinear partial differential equations. SIAM J. Sci. Comput. 40(1) (2018) A172–A198.
[10] Z. Long, Y. Lu, and B. Dong, PDE-Net 2.0: Learning PDEs from data with a numeric-symbolic hybrid deep network, J. Comput. Phys. 399 (2019)108925.
[11] 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, J. Comput. Phys. 378 (2019) 686-707.
[12] G. Pang, L. Lu, and G. E. Karniadakis, PINNs: Fractional physics-informed neural networks, SIAM J. Sci. Comput. 41(4) (2019) A2603–A2626.
[13] H. Xu, H. Chang, and D. Zhang, DLGA-PDE: Discovery of PDEs with incomplete candidate library via combination of deep learning and genetic algorithm, J. Comput. Phys. 418 (2020): 109584.
[14] A.H. Hadian-Rasanan, D. Rahmati, S. Gorgin, and K. Parand, A single layer fractional orthogonal neural network for solving various types of Lane–Emden equation, New Astronomy 75 (2020) 101307.
[15] A.H. Hadian Rasanan, N. Bajalan, K. Parand, and J. Amani Rad, Simulation of nonlinear fractional dynamics arising in the modeling of cognitive decision making using a new fractional neural network, Math. Meth. Appl. Sci. 43(3) (2020) 1437–1466.
[16] A. Gelman, J.B. Carlin, H.S. Stern, D.B. Dunson, A. Vehtari, and D.B Rubin, Bayesian Data Analysis. CRC Press, 2013.
[17] B.M. Turner and T. Van Zandt, A tutorial on approximate Bayesian computation, J.  Math. Psycho. 56(2) (2012) 69–85.
[18] D.B. Rubin, Bayesianly justifiable and relevant frequency calculations for the applied statistician, Ann. Statistics 12(4) (1984) 1151–1172.
[19] S.T. Radev, U.K. Mertens, A. Voss, and U. Kothe, Towards end-to-end likelihood-free inference with convolutional neural networks, Br. J. Math. Statistic. Psycho. 73(1) (2020) 23–43.
[20] K. Diethelm, N.J. Ford, and A.D. Freed, Detailed error analysis for a fractional Adams method, Numerical Algorithms 36(1) (2004) 31–52.
[21] M. Abbaszadeh and M. Dehghan, Fourth-order alternating direction implicit difference scheme to simulate the space-time Riesz tempered fractional diffusion equation, Int. J. Comput. Math. 98(11) (2021) 2137-2155
[22] A. Saadatmandi and M. Dehghan, A new operational matrix for solving fractional-order differential equations, Comput. Math. Appl. 59(3) (2010)1326–1336.
[23] M. Abbaszadeh and H. Amjadian, Second-order finite difference/spectral element formulation for solving the fractional advection-diffusion equation, Commun.Appl. Math. Comput. 2 (2020) 653-669.
[24] M. Yavuz and N. Ozdemir, Numerical inverse Laplace homotopy technique for fractional heat equations, Thermal Sci. 22(2018) 185–194.
[25] J. Eshaghi, H. Adibi, and S. Kazem, On a numerical investigation of the time fractional Fokker–Planck equation via local discontinuous Galerkin method, Int. J. Comput. Math. 94(9) (2017) 1916–1942.
[26] S. Kazem and M. Dehghan, Semi-analytical solution for time-fractional diffusion equation based on finite difference method of lines (MOL), Engin. Comput. 35(1) (2019) 229–241.
Volume 12, Issue 1
May 2021
Pages 189-201
  • Receive Date: 10 May 2020
  • Revise Date: 21 December 2020
  • Accept Date: 26 December 2020