Advertisement

Fast iterative solvers for an optimal transport problem

  • Roland Herzog
  • John W. Pearson
  • Martin Stoll
Article
  • 18 Downloads

Abstract

Optimal transport problems pose many challenges when considering their numerical treatment. We investigate the solution of a PDE-constrained optimisation problem subject to a particular transport equation arising from the modelling of image metamorphosis. We present the nonlinear optimisation problem, and discuss the discretisation and treatment of the nonlinearity via a Gauss–Newton scheme. We then derive preconditioners that can be used to solve the linear systems at the heart of the (Gauss–)Newton method.

Keywords

PDE-constrained optimisation Saddle point systems Time-dependent PDE-constrained optimisation Preconditioning Krylov subspace solver Optical flow Optimal transport 

Mathematics Subject Classification (2010)

65F08 68U10 49J20 35L45 65M06 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Notes

Acknowledgments

The authors would like to thank two anonymous referees for their helpful comments and suggestions. JWP gratefully acknowledges support from the Engineering and Physical Sciences (EPSRC) Fellowship EP/M018857/2.

References

  1. 1.
    Villani, C.: Optimal Transport: Old and New, vol. 338. Springer Science & Business Media, Berlin (2008)Google Scholar
  2. 2.
    Kuzmin, D.: A Guide to Numerical Methods for Transport Equations, Friedrich-Alexander-Universität Erlangen–Nürnberg, http://www.mathematik.uni-dortmund.de/~kuzmin/Transport.pdf (2010)
  3. 3.
    Leveque, R.J.: Numerical Methods for Conservation Laws, 2nd edn. Birkhäuser, Basel (1992)CrossRefMATHGoogle Scholar
  4. 4.
    Ito, K., Kunisch, K.: Lagrange Multiplier Approach to Variational Problems and Applications, vol. 15 of Advances in Design and Control, Society for Industrial and Applied Mathematics. Philadelphia, PA (2008)Google Scholar
  5. 5.
    Tröltzsch, F.: Optimal Control of Partial Differential Equations: Theory, Methods and Applications, American Mathematical Society (2010)Google Scholar
  6. 6.
    Benzi, M., Haber, E., Taralli, L.: A preconditioning technique for a class of PDE-constrained optimization problems. Adv. Comput. Math. 35, 149–173 (2011)MathSciNetCrossRefMATHGoogle Scholar
  7. 7.
    Kantorovich, L.V.: On a problem of Monge. J. Math. Sci. 133(4), 1383–1383 (2006)MathSciNetCrossRefMATHGoogle Scholar
  8. 8.
    Kantorovitch, L.V.: On the translocation of masses. Manag. Sci. 5(1), 1–4 (1958)MathSciNetCrossRefMATHGoogle Scholar
  9. 9.
    Benamou, J. -D., Brenier, Y.: A computational fluid mechanics solution to the Monge–Kantorovich mass transfer problem. Numer. Math. 84(3), 375–393 (2000)MathSciNetCrossRefMATHGoogle Scholar
  10. 10.
    Andreev, R., Scherzer, O., Zulehner, W.: Simultaneous optical flow and source estimation: space time discretization and preconditioning. Appl. Numer. Math. 96, 72–81 (2015)MathSciNetCrossRefMATHGoogle Scholar
  11. 11.
    Borzi, A., Ito, K., Kunisch, K.: Optimal control formulation for determining optical flow. SIAM J. Sci. Comput. 24(3), 818–847 (2003)MathSciNetCrossRefMATHGoogle Scholar
  12. 12.
    Bruhn, A., Weickert, J., Schnörr, C.: Lucas/Kanade meets Horn/Schunck: combining local and global optic flow methods. Int. J. Comput. Vis. 61(3), 211–231 (2005)CrossRefGoogle Scholar
  13. 13.
    Haber, E., Modersitzki, J.: A multilevel method for image registration. SIAM J. Sci. Comput. 27(5), 1594–1607 (2006)MathSciNetCrossRefMATHGoogle Scholar
  14. 14.
    Horn, B.K.P., Schunck, B.G.: Determining optical flow. Artif. Intell. 17 (1–3), 185–203 (1981)CrossRefGoogle Scholar
  15. 15.
    Mang, A., Biros, G.: An inexact Newton–Krylov algorithm for constrained diffeomorphic image registration. SIAM J. Imaging Sci. 8(2), 1030–1069 (2015)MathSciNetCrossRefMATHGoogle Scholar
  16. 16.
    Mang, A., Ruthotto, L.: A Lagrangian Gauss–Newton–Krylov solver for mass- and intensity-preserving diffeomorphic image registration. SIAM J. Sci. Comput. 39 (5), B860–B885 (2017)MathSciNetCrossRefMATHGoogle Scholar
  17. 17.
    Haber, E., Ascher, U.M., Oldenburg, D.: On optimization techniques for solving nonlinear inverse problems. Inverse Probl. 16(5), 1263–1280 (2000)MathSciNetCrossRefMATHGoogle Scholar
  18. 18.
    Axelsson, O., Farouq, S., Neytcheva, M.: Comparison of preconditioned Krylov subspace iteration methods for PDE-constrained optimization problems. Numer. Algorithm. 73(3), 631–663 (2016)CrossRefMATHGoogle Scholar
  19. 19.
    Pearson, J.W., Stoll, M., Wathen, A.J.: Regularization-robust preconditioners for time-dependent PDE-constrained optimization problems. SIAM J. Matrix Anal. Appl. 33(4), 1126–1152 (2012)MathSciNetCrossRefMATHGoogle Scholar
  20. 20.
    Pearson, J.W., Wathen, A.J.: Fast iterative solvers for convection–diffusion control problems. Electron. Trans. Numer. Anal. 40, 294–310 (2013)MathSciNetMATHGoogle Scholar
  21. 21.
    Zulehner, W.: Non-standard norms and robust estimates for saddle point problems. SIAM J. Matrix Anal. Appl. 32, 536–560 (2011)MathSciNetCrossRefMATHGoogle Scholar
  22. 22.
    Nocedal, J., Wright, S.J.: Numerical Optimization, 2nd edn. Springer Series in Operations Research and Financial Engineering, Springer, New York (2006)MATHGoogle Scholar
  23. 23.
    Kuznetsov, Y.A.: Efficient iterative solvers for elliptic finite element problems on nonmatching grids. Russ. J. Numer. Anal. Math. Model. 10, 187–211 (1995)MathSciNetMATHGoogle Scholar
  24. 24.
    Murphy, M.F., Golub, G.H., Wathen, A.J.: A note on preconditioning for indefinite linear systems. SIAM J. Sci. Comput. 21(6), 1969–1972 (2000)MathSciNetCrossRefMATHGoogle Scholar
  25. 25.
    Ipsen, I.C.F.: A note on preconditioning non-symmetric matrices. SIAM J. Sci. Comput. 23(3), 1050–1051 (2001)MathSciNetCrossRefMATHGoogle Scholar
  26. 26.
    Pearson, J.W., Stoll, M.: Fast iterative solution of reaction–diffusion control problems arising from chemical processes. SIAM J. Sci. Comput. 35, B987–B1009 (2013)MathSciNetCrossRefMATHGoogle Scholar
  27. 27.
    Pearson, J.W., Wathen, A.J.: A new approximation of the Schur complement in preconditioners for PDE-constrained optimization. Numer. Linear Algebra Appl. 19, 816–829 (2012)MathSciNetCrossRefMATHGoogle Scholar
  28. 28.
    Benner, P., Dolgov, S., Onwunta, A., Stoll, M.: Low-rank solvers for unsteady Stokes–Brinkman optimal control problem with random data. Comput. Methods Appl. Mech. Eng. 304, 26–54 (2016)MathSciNetCrossRefGoogle Scholar
  29. 29.
    Saad, Y., Schultz, M.H.: GMRES: A generalized minimal residual algorithm for solving nonsymmetric linear systems. SIAM J. Sci. Comput. 7(3), 856–869 (1986)MathSciNetCrossRefMATHGoogle Scholar
  30. 30.
    Borzi, A., Ito, K., Kunisch, K.: An optimal control approach to optical flow computation. Int. J. Numer. Methods Fluids 40(1–2), 231–240 (2002)MathSciNetCrossRefMATHGoogle Scholar
  31. 31.
    Black, M.J., Jepson, A.D.: Eigentracking: robust matching and tracking of articulated objects using a view-based representation. Int. J. Comput. Vis. 26(1), 63–84 (1998)CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Faculty of MathematicsTechnische Universität ChemnitzChemnitzGermany
  2. 2.School of MathematicsThe University of EdinburghEdinburghUK

Personalised recommendations