Monotone Operators and the Proximal Point Algorithm

For the problem of minimizing a lower semicontinuous proper convex function f on a Hilbert space, the proximal point algorithm in exact form generates a sequence $\{ z^k \} $ by taking $z^{k + 1} $ to be the minimizes of $f(z) + ({1 / {2c_k }})\| {z - z^k } \|^2 $, where $c_k > 0$. This algorithm is of interest for several reasons, but especially because of its role in certain computational methods based on duality, such as the Hestenes-Powell method of multipliers in nonlinear programming. It is investigated here in a more general form where the requirement for exact minimization at each iteration is weakened, and the subdifferential $\partial f$ is replaced by an arbitrary maximal monotone operator T. Convergence is established under several criteria amenable to implementation. The rate of convergence is shown to be “typically” linear with an arbitrarily good modulus if $c_k $ stays large enough, in fact superlinear if $c_k \to \infty $. The case of $T = \partial f$ is treated in extra detail. Application is also made to a related case corresponding to minimax problems.

  • [1]  E. Asplund and , R. T. Rockafellar, Gradients of convex functions, Trans. Amer. Math. Soc., 139 (1969), 443–467 MR0240621 0181.41901 CrossrefISIGoogle Scholar

  • [2]  A. Auslender, Problèmes de minimax via l'analyse convexe et les inégalités variationnelles: théorie et algorithmes, Springer-Verlag, Berlin, 1972vii+132, Lecture Notes in Econ. and Math. Systems, 77 MR0461244 0251.90039 Google Scholar

  • [3]  A. B. Bakushinskii and , B. T. Polyak, On the solution of variational inequalities, to appear Google Scholar

  • [4]  Dimitri P. Bertsekas, Necessary and sufficient condition for a penalty method to be exact, Math. Programming, 9 (1975), 87–99 10.1007/BF01681332 MR0384144 0325.90055 CrossrefISIGoogle Scholar

  • [5]  Dimitri P. Bertsekas, Multiplier methods: a survey, Automatica—J. IFAC, 12 (1976), 133–145, March MR0395838 0321.49027 CrossrefISIGoogle Scholar

  • [6]  A. Brøndsted and , R. T. Rockafellar, On the subdifferentiability of convex functions, Proc. Amer. Math. Soc., 16 (1965), 605–611 MR0178103 0141.11801 CrossrefISIGoogle Scholar

  • [7]  Felix E. Browder, Multi-valued monotone nonlinear mappings and duality mappings in Banach spaces, Trans. Amer. Math. Soc., 118 (1965), 338–351 MR0180884 0138.39903 CrossrefISIGoogle Scholar

  • [8]  Magnus R. Hestenes, Multiplier and gradient methods, J. Optimization Theory Appl., 4 (1969), 303–320 10.1007/BF00927673 MR0271809 0174.20705 CrossrefGoogle Scholar

  • [9]  L. V. Kantorovich and , G. P. Akilov, Functional Analysis in Normed Spaces, 1950, English transl., Macmillan, New York, 1964 Google Scholar

  • [10]  M. A. Krasnoselskii˘, Solution of equations involving adjoint operators by successive approximations, Uspehi Mat. Nauk, 15 (1960), 161–165 MR0119100 0095.31504 Google Scholar

  • [11]  A. V. Krjanev, Solution of ill-posed problems by successive approximation methods, Dokl. Akad. Nauk SSSR, 210 (1973), 20–22, Soviet Math. Dokl., 14 (1973), pp. 673–676 MR0322547 0294.47008 Google Scholar

  • [12]  B. Martinet, Régularisation d'inéquations variationnelles par approximations successives, Rev. Française Informat. Recherche Opérationnelle, 4 (1970), 154–158 MR0298899 0215.21103 Google Scholar

  • [13]  Bernard Martinet, Détermination approchée d'un point fixe d'une application pseudo-contractante. Cas de l'application prox, C. R. Acad. Sci. Paris Sér. A-B, 274 (1972), A163–A165 MR0290213 0226.47032 Google Scholar

  • [14]  George J. Minty, Monotone (nonlinear) operators in Hilbert space, Duke Math. J., 29 (1962), 341–346 10.1215/S0012-7094-62-02933-2 MR0169064 0111.31202 CrossrefISIGoogle Scholar

  • [15]  George J. Minty, On the monotonicity of the gradient of a convex function, Pacific J. Math., 14 (1964), 243–247 MR0167859 0123.10601 CrossrefISIGoogle Scholar

  • [16]  Jean-Jacques Moreau, Fonctionnelles sous-différentiables, C. R. Acad. Sci. Paris, 257 (1963), 4117–4119 MR0158250 0137.31401 Google Scholar

  • [17]  Jean-Jacques Moreau, Sur la fonction polaire d'une fonction semi-continue supérieurement, C. R. Acad. Sci. Paris, 258 (1964), 1128–1130 MR0160093 0144.30501 Google Scholar

  • [18]  Jean-Jacques Moreau, Proximitéet dualité dans un espace hilbertien, Bull. Soc. Math. France, 93 (1965), 273–299 MR0201952 0136.12101 CrossrefGoogle Scholar

  • [19]  J. J. Moreau, Fonctionelles Convexes, Seminaire “Equations aux derivées partielles”, lecture notes, Collége de France, Paris, 1966-67 Google Scholar

  • [20]  B. T. Polyak and , N. V. Tretyakov, On an iterative method of linear programming and its economic interpretation, Ekon. Mat. Met., 8 (1972), 740–751 Google Scholar

  • [21]  M. J. D. Powell, R. Fletcher, A method for nonlinear constraints in minimization problemsOptimization (Sympos., Univ. Keele, Keele, 1968), Academic Press, London, 1969, 283–298 MR0272403 0194.47701 Google Scholar

  • [22]  R. T. Rockafellar, Extension of Fenchel's duality theorem for convex functions, Duke Math. J., 33 (1966), 81–89 10.1215/S0012-7094-66-03312-6 MR0187062 0138.09301 CrossrefISIGoogle Scholar

  • [23]  R. T. Rockafellar, Level sets and continuity of conjugate convex functions, Trans. Amer. Math. Soc., 123 (1966), 46–63 MR0192318 0145.15802 CrossrefISIGoogle Scholar

  • [24]  R. T. Rockafellar, F. E. Browder, Monotone operators associated with saddle-functions and minimax problemsNonlinear Functional Analysis (Proc. Sympos. Pure Math., Vol. XVIII, Part 1, Chicago, Ill., 1968), Amer. Math. Soc., Providence, R.I., 1970, 241–250 MR0285942 0237.47030 CrossrefGoogle Scholar

  • [25]  R. T. Rockafellar, Local boundedness of nonlinear, monotone operators, Michigan Math. J., 16 (1969), 397–407 10.1307/mmj/1029000324 MR0253014 0175.45002 CrossrefISIGoogle Scholar

  • [26]  R. Tyrrell Rockafellar, Convex analysis, Princeton Mathematical Series, No. 28, Princeton University Press, Princeton, N.J., 1970xviii+451 MR0274683 0193.18401 CrossrefGoogle Scholar

  • [27]  R. T. Rockafellar, On the maximality of sums of nonlinear monotone operators, Trans. Amer. Math. Soc., 149 (1970), 75–88 MR0282272 0222.47017 CrossrefISIGoogle Scholar

  • [28]  R. T. Rockafellar, H. W. Kuhn and , G. P. Szego, Saddle functions and convex analysisDifferential Games and Related Topics, North-Holland, Amsterdam, 1971, 109–128 0242.90044 Google Scholar

  • [29]  R. T. Rockafellar, The multiplier method of Hestenes and Powell applied to convex programming, J. Optimization Theory Appl., 12 (1973), 555–562 10.1007/BF00934777 MR0334953 0254.90045 CrossrefISIGoogle Scholar

  • [30]  R. Tyrrell Rockafellar, Conjugate duality and optimization, Society for Industrial and Applied Mathematics, Philadelphia, Pa., 1974vi+74, Regional Conference Series in Applied Mathematics No. 16 MR0373611 0296.90036 LinkGoogle Scholar

  • [31]  R. T. Rockafellar, Augmented Lagrangians and applications of the proximal point algorithm in convex programming, Math. Oper. Res., 1 (1976), 97–116 MR0418919 0402.90076 CrossrefGoogle Scholar

  • [32]  Zdzisław Opial, Weak convergence of the sequence of successive approximations for nonexpansive mappings, Bull. Amer. Math. Soc., 73 (1967), 591–597 MR0211301 0179.19902 CrossrefISIGoogle Scholar

  • [33]  A. Genel and , J. Lindenstrauss, An example concerning fixed points, Israel J. Math., 22 (1975), 81–86 MR0390847 0314.47031 CrossrefISIGoogle Scholar