ADMM and Non-convex Variational Problems

Chapter
First Online:
Part of the Scientific Computation book series (SCIENTCOMP)

Abstract

Our main goal in this chapter is to discuss the application of Alternating Direction Methods of Multipliers (ADMM) to the numerical solution of non-convex (and possibly non-smooth) variational problems. After giving a relatively detailed history of the ADMM methodology, we will discuss its application to the solution of problems from nonlinear Continuum Mechanics, nonlinear Elasticity, in particular. The ADMM solution of the two-dimensional Dirichlet problem for the Monge-Ampère equation will be discussed also. The results of numerical experiments will be reported, in order to illustrate the capabilities of the methodology under consideration

This is a preview of subscription content, log in to check access.

Notes

Acknowledgements

The author wants to thank this chapter referee and his present and former colleagues and collaborators J.F. Bourgat, A. Caboussat, E.J. Dean, J.M. Dumay, P. Le Tallec, A. Quaini, T.W. Pan, V. Pons, and L. Tartar for their invaluable help and suggestions. The support of NSF grants DMS 0412267 and DMS 0913982 is also acknowledged.

References

  1. 1.
    Adams, R.A.: Sobolev Spaces. Academic Press, New York, NY (1975)MATHGoogle Scholar
  2. 2.
    Benamou, J.D., Brenier, Y.: A computational fluid mechanics solution to the Monge-Kantorovich mass transfer problem. Numer. Math., 84, 375–393 (2000)CrossRefMATHMathSciNetGoogle Scholar
  3. 3.
    Böhmer, K.: On finite element methods for nonlinear elliptic equations of second order. SIAM J. Numer. Anal., 46, 1212–1249 (2008)CrossRefMATHMathSciNetGoogle Scholar
  4. 4.
    Bourgat, J.F., Dumay, J.M., Glowinski, R.: Large displacement calculations of flexible pipelines by finite element and nonlinear programming methods. SIAM J. Sci. Stat. Comput., 1 (1), 34–81 (1980)CrossRefMATHMathSciNetGoogle Scholar
  5. 5.
    Boyd, S., Parikh, N., Chu, E., Peleato, B., Eckstein, J.: Distributed optimization and statistical learning via the alternating direction method of multipliers. Foundations and Trends in Machine Learning, 3 (1), 1–122 (2011)CrossRefMATHGoogle Scholar
  6. 6.
    Brenier, Y., Frisch, U., Hénon, M., Loeper, G., Mattarese, S., Mohayahee, R., Sobolevskii, A.: Reconstruction of the early Universe as a convex optimization problem. Month. Notices Roy. Astron. Soc.,346 (2), 501–524 (2003)Google Scholar
  7. 7.
    Brenner, S.C., T. Gudi, T., Neilan, M., L.Y. Sung, L.Y.: C 0 penalty methods for the fully nonlinear Monge-Ampère equation. Math. Comp., 80 (276), 1979–1995 (2011)Google Scholar
  8. 8.
    Brenner, S.C., Neilan, M.: Finite element approximations of the three-dimensional Monge-Ampère equation. ESAIM: Math. Model. Numer. Anal., 46 (5), 979–1001 (2012)CrossRefMATHMathSciNetGoogle Scholar
  9. 9.
    Caboussat, A., Glowinski, R., Pons, V.: An augmented Lagrangian approach to the numerical solution of a non-smooth eigenvalue problem. Journal of Numerical Mathematics, 17 (1), 3–26 (2009)CrossRefMATHMathSciNetGoogle Scholar
  10. 10.
    Caboussat, A., Glowinski, R., Sorensen, D.C.: A least-squares method for the numerical solution of the Dirichlet problem for the elliptic Monge-Ampère equation in dimension two. ESAIM: Control Optim. Calcul Variations, 19 (3), 780–810 (2013)CrossRefMATHGoogle Scholar
  11. 11.
    Cabré, X.: Topics in regularity and qualitative properties of solutions of nonlinear elliptic equations. Discr. Cont. Dyn, Syst., 8 (2), 331–360 (2002)Google Scholar
  12. 12.
    Caffarelli, L.A., Cabré, X.: Fully Nonlinear Elliptic Equations. AMS, Providence, RI (1995)CrossRefMATHGoogle Scholar
  13. 13.
    Cea, J., Glowinski, R.: Sur des méthodes d’optimisation par relaxation. ESAIM: Math. Model. Num. Anal., 7 (R3),5–31 (1973)MATHGoogle Scholar
  14. 14.
    Chan, T.F., Glowinski, R.: Finite Element Approximation and Iterative Solution of a Class of Mildly Nonlinear Elliptic Equations. Stanford report STAN-CS-78-674, Computer Science Department, Stanford University, Palo Alto, CA (1978)Google Scholar
  15. 15.
    Chartrand, R.: Non-convex splitting for regularized low-rank + sparse decomposition. IEEE Transactions on Signal Processing, 60 (11), 5810–5819 (2012)CrossRefMathSciNetGoogle Scholar
  16. 16.
    Chartrand, R., Wohlberg, B.: A non-convex ADMM algorithm for group sparsity with sparse groups. In Proceedings of the IEEE 2013 International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 6009–6013, IEEE (2013)Google Scholar
  17. 17.
    Chartrand, R., Sidky, E.Y., Pan, X.: Non-convex compressive sensing for X-ray CT: an algorithm comparison. In Proceedings of the IEEE 2013 Asilomar Conference on Signals, Systems and Computers, pp. 665–669, IEEE (2013)Google Scholar
  18. 18.
    Ciarlet, P.G.: Linear and Nonlinear Functional Analysis with Applications. SIAM, Philadelphia, PA (2013)MATHGoogle Scholar
  19. 19.
    Courant, R., Hilbert, D.: Methods of Mathematical Physics. Volume II: Partial Differential Equations. J. Wiley, New York, NY (1989)CrossRefMATHGoogle Scholar
  20. 20.
    Dean, E.J., Glowinski, R.: Numerical solution of the two-dimensional elliptic Monge-Ampère equation with Dirichlet boundary conditions: an augmented Lagrangian approach. C. R. Math. Acad. Sci. Paris, 336 (9), 779–784 (2003)CrossRefMATHGoogle Scholar
  21. 21.
    Dean, E.J., Glowinski, R.: An augmented Lagrangian approach to the numerical solution of the Dirichlet problem for the elliptic Monge-Ampère equation in two dimensions. Electron. Transact. Num. Anal.,22, 71–96 (2006)Google Scholar
  22. 22.
    Dean, E.J., Glowinski, R.: Numerical methods for fully nonlinear elliptic equations of the Monge-Ampère type. Comp. Meth. Appl. Mech. Eng., 195 (13), 1344–1386 (2006)CrossRefMATHGoogle Scholar
  23. 23.
    Dean, E.J., Glowinski, R., Guidoboni, G.: On the numerical simulation of Bingham visco- plastic flow: Old and new results. J. Non-Newtonian Fluid Mech., 142 (1–3), 36–62 (2007)CrossRefMATHGoogle Scholar
  24. 24.
    Delbos, F., Gilbert, J.C., Glowinski, R., Sinoquet, D.: Constrained optimization in seismic reflection tomography: a Gauss-Newton augmented Lagrangian approach. Geophys. J. International, 164(3), 670–684 (2006)CrossRefGoogle Scholar
  25. 25.
    Douglas, J., Rachford, H.H.: On the solution of the heat conduction problem in 2 and 3 space variables. Trans. Amer. Math. Soc., 82, 421–439 (1956)CrossRefMATHMathSciNetGoogle Scholar
  26. 26.
    Duvaut, G., Lions, J.L.: Inequalities in Mechanics and Physics. Springer, Berlin (1976)CrossRefMATHGoogle Scholar
  27. 27.
    Ekeland, I., Temam, R.: Convex Analysis and Variational Problems. SIAM, Philadelphia, PA (1999)CrossRefMATHGoogle Scholar
  28. 28.
    Feng, X., Glowinski, R., Neilan, M.: Recent developments in numerical methods for fully nonlinear partial differential equations. SIAM Rev., 55 (2), 205–267 (2013)CrossRefMATHMathSciNetGoogle Scholar
  29. 29.
    Fortin, M., Glowinski, R.: Méthodes de Lagrangiens Augmentés: Application à la Résolution Numérique des Problèmes aux Limites. Dunod, Paris (1982)Google Scholar
  30. 30.
    Fortin, M., Glowinski, R.: Augmented Lagrangian Methods: Applications to the Numerical Solution of Boundary Value Problems. North-Holland, Amsterdam (1983)MATHGoogle Scholar
  31. 31.
    Glowinski, R.: Numerical Methods for Nonlinear Variational Problems. Springer, New York NY (1984) (2nd printing: 2008)CrossRefMATHGoogle Scholar
  32. 32.
    Glowinski, R.: Finite element methods for incompressible viscous flow. In: Ciarlet, P.G., Lions, J.L. (eds.) Handbook of Numerical Analysis, Vol. IX, pp. 3–1176. North-Holland, Amsterdam (2003)Google Scholar
  33. 33.
    Glowinski, R.: On alternating direction methods of multipliers: a historical perspective. In Fitzgibbon, W., Kuznetsov, Y.A., Neittannmäki, P., Pironneau, O. (eds.) Modeling, Simulation and Optimization for Science and Technology, pp. 52–82. Springer, Dordrecht (2014)Google Scholar
  34. 34.
    Glowinski, R.: Variational Methods for the Numerical Solution of Nonlinear Elliptic Problems. SIAM, Philadelphia, PA (2015).CrossRefMATHGoogle Scholar
  35. 35.
    Glowinski, R., Dean, E.J., Guidoboni, G., Juarez, H.L., Pan, T.W.: Applications of operator-splitting methods to the direct numerical simulation of particulate and free surface flows and to the numerical solution of the two-dimensional elliptic Monge-Ampère equation. Japan J. Ind. Appl. Math., 25 (1), 1–63 (2008)CrossRefMATHGoogle Scholar
  36. 36.
    Glowinski, R., Hölmström, M.: Constrained motion problems with applications by nonlinear programming methods. Survey on Mathematics for Industry, 5, 75–108 (1995)MATHMathSciNetGoogle Scholar
  37. 37.
    Glowinski, R., Le Tallec, P.: Numerical solution of problems in incompressible finite elasticity by augmented Lagrangian methods. I. Two-dimensional and axisymmetric problems. SIAM J. Appl. Math., 42 (2), 400–429 (1982)MATHGoogle Scholar
  38. 38.
    Glowinski, R., Le Tallec, P.: Numerical solution of problems in incompressible finite elasticity by augmented Lagrangian methods. II. Three-dimensional problems. SIAM J. Appl. Math., 44 (4), 710–733 (1984)CrossRefMATHGoogle Scholar
  39. 39.
    Glowinski, R., Le Tallec, P.: Augmented Lagrangian and Operator-Splitting Methods in Nonlinear Mechanics. SIAM, Philadelphia, PA (1989)CrossRefMATHGoogle Scholar
  40. 40.
    Glowinski, R., Lions, J.L., Trémolières, R.: Numerical Analysis of Variational Inequalities. North-Holland, Amsterdam (1981)MATHGoogle Scholar
  41. 41.
    Glowinski, R., Marrocco, A.: Sur l’approximation par éléments finis d’ordre un et la résolution par pénalisation-dualité d’une classe de problèmes de Dirichlet non linéaires. C.R. Acad. Sci. Paris, 278A, 1649–1652 (1974)MATHGoogle Scholar
  42. 42.
    Glowinski, R., Marrocco, A.: Sur l’approximation par éléments finis d’ordre un et la résolution par pénalisation-dualité d’une classe de problèmes de Dirichlet non linéaires. ESAIM: Math. Model. Num. Anal., 9 (R2), 41–76 (1975)MATHGoogle Scholar
  43. 43.
    Glowinski, R., Quaini, A.: On an inequality of C. Sundberg: A computational investigation via nonlinear programming. J. Optim. Theory Appl., 158 (3), 739–772 (2013)MATHGoogle Scholar
  44. 44.
    Glowinski, R., Wachs, A.: On the numerical simulation of visco-plastic fluid flow. In: Ciarlet, P.G., Glowinski, R., Xu, J. (eds.) Handbook of Numerical Analysis, Vol. XVI, North-Holland, Amsterdam, pp. 483–717 (2011)Google Scholar
  45. 45.
    Guttiérez, C.: The Monge-Ampère Equation. Birkhäuser, Boston, MA (2001)CrossRefGoogle Scholar
  46. 46.
    He, J.W., Glowinski, R.: Steady Bingham fluid flow in cylindrical pipes: a time dependent approach to the iterative solution. Num. Linear Algebra Appl., 7 (6), 381–428 (2000)CrossRefMATHMathSciNetGoogle Scholar
  47. 47.
    Ito, K., Kunish, K.: Lagrange Multiplier Approach to Variational Problems and Applications. SIAM, Philadelphia, PA (2008)CrossRefGoogle Scholar
  48. 48.
    Lagnese, J., Lions, J.L.: Modelling, Analysis and Control of Thin Plates. Masson, Paris (1988)MATHGoogle Scholar
  49. 49.
    Le Tallec, P.: Numerical methods for nonlinear three-dimensional elasticity. In: Ciarlet, P.G., Lions, J.L., (eds.) Handbook of Numerical Analysis, Vol. 3, North-Holland, Amsterdam, pp. 465–622 (1994)Google Scholar
  50. 50.
    Lions, J.L.: Optimal Control of Systems Governed by Partial Differential Equations. Springer, Berlin (1971)CrossRefMATHGoogle Scholar
  51. 51.
    Majava, K., Glowinski, R., Kärkkäinen, T.: Solving a non-smooth eigenvalue problem using operator-splitting methods. Inter. J. Comp. Math., 84 (6), 825–846 (2007)CrossRefMATHMathSciNetGoogle Scholar
  52. 52.
    Nečas, J.: Les Méthodes Directes en Théorie des Equations Elliptiques. Masson, Paris (1967)Google Scholar
  53. 53.
    Nečas, J.: Direct Methods in the Theory of Elliptic Equations. Springer, Heidelberg (2011)Google Scholar
  54. 54.
    Neilan, M.: A nonconforming Morley finite element method for the fully nonlinear Monge-Ampère equation. Numer. Math., 115 (3), 371–394 (2010)CrossRefMATHMathSciNetGoogle Scholar
  55. 55.
    Oliker, V., Prussner, L.: On the numerical solution of the equation z xx z yy − z xy = f and its discretization. I, Numer. Math., 54 (3), 271–293 (1988)Google Scholar
  56. 56.
    Peaceman, D.H., Rachford, H.H.: The numerical solution of parabolic and elliptic differential equations. J. SIAM, 3, 28–41 (1955)MATHMathSciNetGoogle Scholar
  57. 57.
    Raviart, P.A., Thomas, J.M.: Introduction à l’Analyse Numérique des Equations aux Dérivées Partielles. Masson, Paris (1983)MATHGoogle Scholar
  58. 58.
    Schäfer, M.: Parallel algorithms for the numerical solution of incompressible finite elasticity problems. SIAM J. Sci. Stat. Comput., 12, 247–259 (1991)CrossRefMATHMathSciNetGoogle Scholar
  59. 59.
    Schwartz, L.: Théorie des Distributions. Hermann, Paris (1966)MATHGoogle Scholar
  60. 60.
    Sorensen, D.C., Glowinski, R.: A quadratically constrained minimization problem arising from PDE of Monge-Ampère type. Numer. Algor.. 53 (1), 53–66 (2010)CrossRefMATHGoogle Scholar
  61. 61.
    Talenti, G.: Best constant in Sobolev inequality. Ann. Mat. Pura Appl., 110 (1), 353–372 (1976)CrossRefMATHMathSciNetGoogle Scholar
  62. 62.
    Tartar, L.: An Introduction to Sobolev Spaces and Interpolation Spaces. Springer, Berlin (2007)MATHGoogle Scholar
  63. 63.
    Villani, C.: Topics in Optimal Transportation. AMS, Providence, RI (2003)CrossRefMATHGoogle Scholar
  64. 64.
    Wang, Y., Yin, W., Zeng, J.: Global convergence of ADMM in nonconvex nonsmooth optimization. arXiv:1511.06324 [cs, math] (2015)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  1. 1.Department of MathematicsUniversity of HoustonHoustonUSA

Personalised recommendations