La méthode de régularisation de Tikhonov pour les problèmes non linéaires mal posés requiert une solution optimale globale des problèmes d'optimisation non convexe qui ont été très peu étudiés dans la communauté des problèmes inverses. Dans ce papier nous suggérons une méthode qui est applicable à une large classe des problèmes non linéaires mal posés. C'est une classe de problèmes dans lesquels la fonctionnelle de Tikhonov peut être représentée comme différences de fonctionnelles convexes (dc). Notre méthode pour ces problèmes est une combinaison de l'algorithme DCA, récemment développé en optimisation dc, et les techniques de séparation et évaluation.
The Tikhonov regularization method for non-linear ill-posed problems requires us to globally solve non-convex optimization problem which have been very little studied in the inverse problems community. In this paper we suggest a method which is applicable to the Tikhonov method for a wide class of non-linear ill-posed problems. This is a class of problems when the Tikhonov functional for them can be represented by the difference of two convex functionals. Our method for these problems is a combination of the recently developed algorithm DCA in dc programming with the branch-and-bound techniques.
Accepté le :
Publié le :
@article{CRMATH_2002__335_12_1073_0, author = {Le Thi Hoai An and Pham Dinh Tao and Dinh Nho H\`ao}, title = {Towards {Tikhonov} regularization of non-linear ill-posed problems: a dc programming approach}, journal = {Comptes Rendus. Math\'ematique}, pages = {1073--1078}, publisher = {Elsevier}, volume = {335}, number = {12}, year = {2002}, doi = {10.1016/S1631-073X(02)02611-0}, language = {en}, url = {http://archive.numdam.org/articles/10.1016/S1631-073X(02)02611-0/} }
TY - JOUR AU - Le Thi Hoai An AU - Pham Dinh Tao AU - Dinh Nho Hào TI - Towards Tikhonov regularization of non-linear ill-posed problems: a dc programming approach JO - Comptes Rendus. Mathématique PY - 2002 SP - 1073 EP - 1078 VL - 335 IS - 12 PB - Elsevier UR - http://archive.numdam.org/articles/10.1016/S1631-073X(02)02611-0/ DO - 10.1016/S1631-073X(02)02611-0 LA - en ID - CRMATH_2002__335_12_1073_0 ER -
%0 Journal Article %A Le Thi Hoai An %A Pham Dinh Tao %A Dinh Nho Hào %T Towards Tikhonov regularization of non-linear ill-posed problems: a dc programming approach %J Comptes Rendus. Mathématique %D 2002 %P 1073-1078 %V 335 %N 12 %I Elsevier %U http://archive.numdam.org/articles/10.1016/S1631-073X(02)02611-0/ %R 10.1016/S1631-073X(02)02611-0 %G en %F CRMATH_2002__335_12_1073_0
Le Thi Hoai An; Pham Dinh Tao; Dinh Nho Hào. Towards Tikhonov regularization of non-linear ill-posed problems: a dc programming approach. Comptes Rendus. Mathématique, Tome 335 (2002) no. 12, pp. 1073-1078. doi : 10.1016/S1631-073X(02)02611-0. http://archive.numdam.org/articles/10.1016/S1631-073X(02)02611-0/
[1] Convergence of Tikhonov regularization for constrained ill-posed inverse problems, Inverse Problems, Volume 10 (1994), pp. 63-76
[2] Regularization of Inverse Problems, Kluwer Academic, Dordrecht, 1996
[3] V. Dicken, Simultaneous activity and attennuation reconstruction in single photon emission computed tomography, a non-linear ill-posed problem, Ph.D. thesis, Universität Potsdam, 1997
[4] Methods for Inverse Heat Conduction Problems, Peter Lang, Frankfurt am Main, 1998
[5] Le Thi Hoai An, Contribution à l'optimisation non convexe et l'optimisation globale : Théorie, algorithmes et applications, Habilitation à Diriger des Recherches, Université de Rouen, 1997
[6] Le Thi Hoai An, Pham Dinh Tao, Dinh Nho Hào, D.c. programming approach to Tikhonov regularization for non-linear ill-posed problems, Preprint, 2001
[7] Le Thi Hoai An, Pham Dinh Tao, Dinh Nho Hào, Solving an inverse problem for an elliptic equation by d.c. programming. J. Global Optimization, 2001, to appear
[8] Algorithms for solving a class of non convex optimization problems. Methods of subgradients, Fermat Days 85. Mathematics for Optimization, Elsevier, North-Holland, 1986, pp. 249-270
[9] Duality in d.c. (difference of convex functions) optimization. Subgradient methods, Trends in Mathematical Optimization, Internat. Ser. Numer Math, 84, Birkhäuser, 1988, pp. 277-293
[10] D.c. optimization algorithms for solving the trust region subproblem, SIAM J. Optimization, Volume 8 (1998), pp. 476-505
[11] Convex analysis approach to d.c. programming: Theory, Algorithms and Applications, Acta Math. Vietnam, Volume 22 (1997) no. 1, pp. 289-355
[12] Le Thi Hoai An, Pham Dinh Tao, Large scale global molecular optimization from exact distance matrices by a d.c. optimization approach, revised version in SIAM J. Optim., 2002
[13] Le Thi Hoai An, Pham Dinh Tao, The dc (difference of convex functions) programming and DCA revisited with dc models of real world nonconvex optimization problems, 2002, submitted
[14] Non-Linear Ill-Posed Problems, Vols. 1, 2, Chapman and Hall, London, 1998
Cité par Sources :