Nonlinear model reduction on metric spaces. Application to one-dimensional conservative PDEs in Wasserstein spaces

Ehrlacher, Virginie; Lombardi, Damiano; Mula, Olga; Vialard, François-Xavier

ESAIM: Mathematical Modelling and Numerical Analysis , Tome 54 (2020) no. 6, pp. 2159-2197.

Suite au passage du modèle économique de la revue en S20, le texte intégral des articles des années 2019 et 2020 est accessible uniquement sur le site de l'éditeur et est réservé aux abonnés.

We consider the problem of model reduction of parametrized PDEs where the goal is to approximate any function belonging to the set of solutions at a reduced computational cost. For this, the bottom line of most strategies has so far been based on the approximation of the solution set by linear spaces on Hilbert or Banach spaces. This approach can be expected to be successful only when the Kolmogorov width of the set decays fast. While this is the case on certain parabolic or elliptic problems, most transport-dominated problems are expected to present a slow decaying width and require to study nonlinear approximation methods. In this work, we propose to address the reduction problem from the perspective of general metric spaces with a suitably defined notion of distance. We develop and compare two different approaches, one based on barycenters and another one using tangent spaces when the metric space has an additional Riemannian structure. Since the notion of linear vectorial spaces does not exist in general metric spaces, both approaches result in nonlinear approximation methods. We give theoretical and numerical evidence of their efficiency to reduce complexity for one-dimensional conservative PDEs where the underlying metric space can be chosen to be the L²-Wasserstein space.

[1] R. Abgrall, D. Amsallem and R. Crisovan, Robust model reduction by $L^1$-norm minimization and approximation via dictionaries: application to nonlinear hyperbolic problems. Adv. Model. Simul. Eng. Sci. 3 (2016) 1. | DOI

[2] B.M. Afkham and J.S. Hesthaven, Structure preserving model reduction of parametric hamiltonian systems. SIAM J. Sci. Comput. 39 (2017) A2616–A2644. | DOI | MR

[3] M. Agueh and G. Carlier, Barycenters in the Wasserstein space. SIAM J. Math. Anal. 43 (2011) 904–924. | DOI | MR | Zbl

[4] D. Amsallem and C. Farhat, Interpolation method for adapting reduced-order models and application to aeroelasticity. AIAA J. 46 (2008) 1803–1813. | DOI

[5] D. Amsallem and B. Haasdonk, PEBL-ROM: projection-error based local reduced-order models. Adv. Model. Simul. Eng. Sci. 3 (2016) 6. | DOI

[6] D. Amsallem, M.J. Zahr and C. Farhat, Nonlinear model order reduction based on local reduced-order bases. Int. J. Numer. Methods Eng. 92 (2012) 891–916. | DOI | MR

[7] M. Barrault, Y. Maday, N.C. Nguyen and A.T. Patera, An “empirical interpolation” method: application to efficient reduced-basis discretization of partial differential equations. C. R. Acad. Sci. Paris Sér. I 339 (2004) 667–672. | DOI | MR | Zbl

[8] P. Benner, A. Cohen, M. Ohlberger and K. Willcox, Model Reduction and Approximation: Theory and Algorithms. In Vol. 15. SIAM (2017). | DOI | MR

[9] J. Bigot, R. Gouet, T. Klein and A. López, Geodesic PCA in the Wasserstein space by convex PCA. Ann. Inst. Henri Poincaré, Proba. Stat. 53 (2017) 1–26. | MR

[10] P. Binev, A. Cohen, W. Dahmen, R. Devore, G. Petrova and P. Wojtaszczyk, Convergence rates for greedy algorithms in reduced basis methods. SIAM J. Math. Anal. 43 (2011) 1457–1472. | DOI | MR | Zbl

[11] P. Binev, A. Cohen, O. Mula and J. Nichols, Greedy algorithms for optimal measurements selection in state estimation using reduced models. SIAM/ASA J. Uncertainty Quant. 6 (2018) 1101–1126. | DOI | MR

[12] A. Blanchet and P. Laurençot, The parabolic-parabolic Keller-Segel system with critical diffusion as a gradient flow in ${ℝ}^d,d\ge 3$., Commun. Part. Differ. Equ. 38 (2013) 658–686. | DOI | MR | Zbl

[13] A. Bressan and M. Fonte, An optimal transportation metric for solutions of the Camassa-Holm equation. Methods Appl. Anal. 12 (2005) 191–219. | DOI | MR | Zbl

[14] N. Cagniart, Y. Maday and B. Stamm, Model order reduction for problems with large convection effects. In: Contributions to Partial Differential Equations and Applications. Springer (2019) 131–150. | DOI | MR

[15] K. Carlberg, Adaptive $h$-refinement for reduced-order models. Int. J. Numer. Methods Eng. 102 (2015) 1192–1210. | DOI | MR

[16] J.A. Carrillo, K. Grunert and H. Holden, A Lipschitz metric for the Hunter-Saxton equation. Commun. Part. Diff. Equ. 44 (2019) 309–334. | DOI | MR

[17] E. Cazelles, V. Seguy, J. Bigot, M. Cuturi and N. Papadakis, Log-PCA versus Geodesic PCA of histograms in the Wasserstein space. Preprint (2017). | arXiv | MR

[18] L. Chizat, G. Peyré, B. Schmitzer and F.-X. Vialard, Scaling algorithms for unbalanced transport problems. Math. Comput. 87 (2018) 2563–2609. | DOI | MR

[19] L. Chizat, B. Schmitzer, G. Peyré and F.-X. Vialard, An interpolating distance between optimal transport and Fischer-Rao. Found. Comput. Math. 18 (2018) 1–44. | DOI | MR

[20] A. Cohen and R. Devore, Kolmogorov widths under holomorphic mappings. IMA J. Numer. Anal. 36 (2016) 1–12. | MR

[21] A. Cohen, R. Devore and C. Schwab, Convergence rates of best $n$-term galerkin approximations for a class of elliptic sPDEs. Found. Comput. Math. 10 (2010) 615–646. | DOI | MR | Zbl

[22] A. Cohen, R. Devore and C. Schwab, Analytic regularity and polynomial approximation of parametric and stochastic elliptic PDE’s. Anal. App. 9 (2011) 11–47. | DOI | MR | Zbl

[23] A. Cohen, W. Dahmen and R. Devore, Reduced basis greedy selection using random training sets. Preprint (2018). | arXiv | Numdam | MR

[24] Z. Ding, G. Fleishman, X. Yang, P. Thompson, R. Kwitt and M. Niethammer, Fast predictive simple geodesic regression. In: 14th International Conference, edited by M. Cardoso, et al. Vol. 1053 of Lecture Notes in Computer Science. Springer, Cham (2017).

[25] F. Feppon and P.F.J. Lermusiaux, A geometric approach to dynamical model order reduction. SIAM J. Matrix Anal. App. 39 (2018) 510–538. | DOI | MR

[26] P.T. Fletcher, C. Lu, S.M. Pizer and S.C. Joshi, Principal geodesic analysis for the study of nonlinear statistics of shape. IEEE Trans. Med. Imaging 23 (2004) 995–1005. | DOI

[27] J.P. Gazeau and P. Winternitz, Symmetries of variable coefficient Korteweg–de Vries equations. J. Math. Phys. 33 (1992) 4087–4102. | DOI | MR | Zbl

[28] L. Giacomelli and F. Otto, Variatonal formulation for the lubrication approximation of the Hele-Shaw flow. Calc. Var. Part. Diff. Equ. 13 (2001) 377–403. | DOI | MR | Zbl

[29] U. Gianazza, G. Savaré and G. Toscani, The Wasserstein gradient flow of the Fisher information and the quantum drift-diffusion equation. Arch. Ratio. Mech. Anal. 194 (2009) 133–220. | DOI | MR | Zbl

[30] F.J. Gonzalez and M. Balajewicz, Learning low-dimensional feature dynamics using deep convolutional recurrent autoencoders. Preprint (2018). | arXiv

[31] C. Greif and K. Urban, Decay of the Kolmogorov $N$-width for wave problems. Appl. Math. Lett. 96 (2019) 216–222. | DOI | MR

[32] M.A. Grepl, Y. Maday, N.C. Nguyen and A.T. Patera, Efficient reduced-basis treatment of nonaffine and nonlinear partial differential equations. ESAIM: M2AN 41 (2007) 575–605. | DOI | Numdam | MR | Zbl

[33] J.S. Hesthaven and C. Pagliantini, Structure-preserving reduced basis methods for hamiltonian systems with a nonlinear poisson structure. Technical report (2018).

[34] J.S. Hesthaven, G. Rozza and B. Stamm, Certified Reduced Basis Methods for Parametrized Partial Differential Equations. SpringerBriefs in Mathematics (2015). | MR

[35] S. Huckemann, T. Hotz and A. Munk, Intrinsic shape analysis: geodesic PCA for Riemannian manifolds modulo isometric Lie group actions. Stat. Sin. 20 (2010) 1–58. | MR | Zbl

[36] A. Iollo and D. Lombardi, Advection modes by optimal mass transfer. Phys. Rev. E 89 (2014) 022923. | DOI

[37] R. Jordan, D. Kinderlehrer and F. Otto, The variational formulation of the Fokker-Planck equation. SIAM J. Math. Anal. 29 (1998) 1–17. | DOI | MR | Zbl

[38] B.N. Khoromskij and C. Schwab, Tensor-structured Galerkin approximation of parametric and stochastic elliptic PDEs. SIAM J. Sci. Comput. 33 (2011) 364–385. | DOI | MR | Zbl

[39] O. Koch and C. Lubich, Dynamical low-rank approximation. SIAM J. Matrix Anal. App. 29 (2007) 434–454. | DOI | MR | Zbl

[40] O. Koch and C. Lubich, Dynamical tensor approximation. SIAM J. Matrix Anal. App. 31 (2010) 2360–2375. | DOI | MR | Zbl

[41] K. Lee and K. Carlberg, Model reduction of dynamical systems on nonlinear manifolds using deep convolutional autoencoders. Preprint (2018). | arXiv | MR

[42] M. Liero, A. Mielke and G. Savaré, Optimal transport in competition with reaction: the Hellinger-Kantorovich distance and geodesic curves. SIAM J. Math. Anal. 48 (2016) 2869–2911. | DOI | MR

[43] Y. Maday and O. Mula, A Generalized Empirical Interpolation Method: application of reduced basis techniques to data assimilation, edited by F. Brezzi, P.C. Franzone, U. Gianazza and G. Gilardi. In: Vol. 4 of Springer INdAM Series. Analysis and Numerics of Partial Differential Equations, Springer Milan (2013) 221–235. | DOI | MR | Zbl

[44] Y. Maday, A. Manzoni and A. Quarteroni, An online intrinsic stabilization strategy for the reduced basis approximation of parametrized advection-dominated problems. C. R. Math. 354 (2016) 1188–1194. | DOI | MR

[45] Y. Maday, O. Mula and G. Turinici, Convergence analysis of the generalized empirical interpolation method. SIAM J. Numer. Anal. 54 (2016) 1713–1731. | DOI | MR

[46] R. Mosquera, A. Hamdouni, A. El Hamiidi and C. Allery, POD basis interpolation via Inverse Distance Weighting on Grassmann manifolds. Disc. Cont. Dyn. Sys. – S 12 (2019) 1743. | MR

[47] S. Mowlavi and T.P. Sapsis, Model order reduction for stochastic dynamical systems with continuous symmetries. SIAM J. Sci. Comput. 40 (2018) A1669–A1695. | DOI | MR

[48] E. Musharbash, F. Nobile and T. Zhou, Error analysis of the dynamically orthogonal approximation of time dependent random PDEs. SIAM J. Sci. Comput. 37 (2015) A776–A810. | DOI | MR

[49] N.J. Nair and M. Balajewicz, Transported snapshot model order reduction approach for parametric, steady-state fluid flows containing parameter-dependent shocks, Int. J. Numer. Methods Eng. (2018) 117 (2019) 1234–1262. | DOI | MR

[50] M. Niethammer, R. Kwitt and F.-X. Vialard, Metric learning for image registration. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2019) 8463–8472.

[51] M. Ohlberger and S. Rave, Nonlinear reduced basis approximation of parameterized evolution equations via the method of freezing. C. R. Math. 351 (2013) 901–906. | DOI | MR | Zbl

[52] M. Ohlberger and S. Rave, Reduced basis methods: success, limitations and future challenges. In: Proceedings of the Conference Algoritmy (2016) 1–12.

[53] F. Otto, The geometry of dissipative evolution equations: the porous medium equation. Commun. Part. Differ. Equ. 26 (2001) 101–174. | DOI | MR | Zbl

[54] B. Peherstorfer, D. Butnaru, K. Willcox and H.-J. Bungartz, Localized discrete empirical interpolation method. SIAM J. Sci. Comput. 36 (2014) A168–A192. | DOI | MR | Zbl

[55] X. Pennec, Barycentric subspace analysis on manifolds. Ann. Stat. 46 (2018) 2711–2746. | DOI | MR | Zbl

[56] A. Quarteroni, A. Manzoni and F. Negri, Reduced Basis Methods for Partial Differential Equations: An Introduction. Springer 92 (2015). | MR | Zbl

[57] Z. Shen, F.-X. Vialard and M. Niethammer, Region-specific diffeomorphic metric mapping. NeurIPS. Preprint (2019). | arXiv

[58] S. Sommer, F. Lauze and M. Nielsen, Optimization over geodesics for exact principal geodesic analysis. Adv. Comput. Math. 40 (2014) 283–313. | DOI | MR | Zbl

[59] D. Torlo, F. Ballarin and G. Rozza, Stabilized weighted reduced basis methods for parametrized advection dominated problems with random inputs. SIAM/ASA J. Uncertainty Quant. 6 (2018) 1475–1502. | DOI | MR | Zbl

[60] C. Villani, Topics in Optimal Transportation. In: Vol. 58 of Graduate Studies in Mathematics. American Mathematical Society (2003). | MR | Zbl

[61] G. Welper, Transformed snapshot interpolation. Preprint (2015). | arXiv

[62] G. Welper, h and hp-adaptive interpolation by transformed snapshots for parametric and stochastic hyperbolic PDEs. Preprint (2017). | arXiv

[63] J. Zinsl and D. Matthes, Exponential convergence to equilibrium in a coupled gradient flow system modeling chemotaxis. Anal. Partial Differ. Equ. 8 (2015) 425–466. | MR | Zbl