1. Intuitive background. Statement of the problem...................................................................... 5 2. General structure of global stochastic approximation processes............................................... 7 3. The fundamental theorem on convergence in distribution............................................................ 10 4. Absolute continuity of the limit distribution 4.1. Introductory remarks............................................................................................................. 13 4.2. General case......................................................................................................................... 13 4.3. Uniform experimental design............................................................................................. 14 4.4. Improvement by a randomization....................................................................................... 16 4.5. Problem of optimal experimental design......................................................................... 19 5. Almost sure convergence to global maximum................................................................................ 21 6. A Monte Carlo method.......................................................................................................................... 24 References................................................................................................................................................. 26
[1] S. H. Brooks, A discussion of random, methods for seeking maxima, Opns. 6 (1958) pp. 244-251.
[2] J. L. Doob, Stochastic processes, Wiley and Sons 1953.
[3] M. Driml and O. Hanš, On a randomised optimization procedure, Trans. 4-th Praque Conf. on Information Theory, Statistical Decision Functions, Random Processes, Prague 1967.
[4] V. Fabian, Stochastic approximation of minima with improved asymptotic speed, Ann. Math. Statist. 38 (1967), pp. 191-200.
[5] V. Fabian, Stochastic approximation, In: Optimising methods in statistics, Ed. J. S. Rustagi. Academic Press, 1971.
[6] Б. П. Харламов, Об одном алгоритме стохастического поиска максимума в детерминированном поле, Труды матем. института им. В. А. Стеклова LXXIX (1965), pp. 71-75.
[7] С. С. Hey de, On martingale limit theory and strong convergence results for stochastic approximation procedures, Stochastic Processes and their Applications 2 (1974), pp. 359-370.
[8] S. M. Jermakow, Metoda Monte Carlo i zagadnienia pokrewne, PWN, Warszawa 1976.
[9] В. Г. Карманов, О сходимости метода случайного поиска в выпуклых задачах минимизации, Теория вероят. и ее примен., XIX, 4 (1974) pp. 817-824.
[10] J. Kiefer and J. Wolfowitz, Stochastic estimation of the maximum of a regression function, Ann. Math. Statist. 23 (1952), pp. 462-466.
[11] P. Major, A law of the iterated logarithm for the Robbins-Monro method, Stud, sci. math. Hung. 8, 1-2 (1973), pp. 95-102.
[12] J. von Neumann, Various technique used in connection with random digits, Nat. Bur. Stand. Appl. Math. Ser. 12 (1951), pp. 36-38.
[13] В. Т. Перекрест, Об одной адаптивной схеме глобального поиска, УМН 29, 3 (1974), pp. 223-224.
[14] Л. А. Растригин, Системы экстремального управления, "Наука", Москва 1974.
[15] Н. Robbins and S. Monro, A stochastic approximation method, Ann. Math. Statist. 22 (1951), pp. 400-407.
[16] Э.М Вайсборд., Д.Б. Юдин, Многоэкстремальная стохастическая аппроксимация, Изв. АН СССР, Сер. техн. кибернетика. 5 (1968), pp. 3-13.
[17] Э.М Вайсборд., Д.Б. Юдин, Стохастическая аппроксимация для многоэкстремальных задач в гильбертовом пространнстве, ДАН СССР 181, 5 (1968), pp. 1034—1037.
[18] М. Т. Wasan, Stochastic approximation, Cambridge 1969.
[19] R. Zieliński, On convergence of a randomized optimisation procedure, Algorytmy 7, 12 (1970), pp. 29-32.
[20] R. Zieliński, A Monte Carlo estimation of the maximum of a function, ibidem 7, 13 (1970), pp. 5-7.
[21] R. Zieliński, A randomized Kiefer-Wolfowitz procedure, European Meeting of Statisticians and 7-th Prague Conference on Information Theory, Statistical Decision Functions and Random Processes, Prague