Les algorithmes stochastiques contournent-ils les pièges ?

Brandière, Odile; Duflo, Marie

Annales de l'I.H.P. Probabilités et statistiques, Tome 32 (1996) no. 3, pp. 395-427

[1] W.J. Barlow, Coefficient properties of random variable sequences. Ann. of probability, vol. 3, 1975, p. 840-848. | Zbl | MR

[2] A Benveniste, M. Métivier et P. Priouret, Algorithmes adaptatifs et approximations stochastiques. Masson, 1987.

[3] O. Brandière, Un algorithme du gradient pour l'analyse en composantes principales. Comptes Rendus Académie des Sciences, 321, série I, 1995, p. 233-236. | Zbl | MR

[4] D.L. Burkholder, Independent sequences with the Stein property, Ann. of math. stat., vol. 39, 1968, p. 1282-1288. | Zbl | MR

[5] B. Delyon, A deterministic approach to stochastic approximation, IRISA, prépublication n° 789, 1994.

[6] D.P. Derevitskii et A.L. Fradkov, Two models for analysing the dynamics of adaption algorithms. Avtomatika i Telemekhanika, 1, 1974, p. 67-75; en anglais, Automation and remote control, vol. 1, 1974, p. 59-67. | Zbl | MR

[7] M. Duflo, Méthodes récursives aléatoires. Masson, 1990. | Zbl | MR

[8] J.C. Fort et G. Pagès, Sur la convergence presque sûre d'algorithmes stochastiques : le théorème de Kushner-Clark revisité. Prépublication du SAMOS 33, Université Paris 1, 1994.

[9] S.B. Gelfand et S.K. Mitter, Recursive stochastic algorithms for global optimization in Rd. SIAM J. control and optimization, vol. 29, 1991, p. 999-1018. | Zbl | MR

[10] S.B. Gelfand et S.K. Mitter, Metropolis-type annealing algorithms for global optimization in Rd. SIAM J. control and optimization, vol. 31, 1993, p. 111-131. | Zbl | MR

[11] P. Hartman, Ordinary Differential Equations, Wiley, 1964; seconde édition, 1982. | Zbl

[12] K. Hornik et C.M. Kuan, Convergence analysis of local feature extraction algorithms. Neural networks, vol. 5, 1992, p. 229-240.

[13] C.R. Hwang et S.J. Sheu, On the behaviour of a stochastic algorithm with annealing. Rapport technique, Academia sinica, Taiwan, 1990.

[14] B. Jessen et A. Wintner, Distribution functions and the Riemann zeta function. Trans. american math. soc., vol. 38, 1935, p. 725-734. | Zbl | MR

[15] J. Kiefer et J. Wolfowitz, Stochastic estimation of the maximum of a regression fonction. Ann. math. stat., vol. 23, 1952, p. 462-466. | Zbl | MR

[16] H.J. Kushner, Asymptotic global behavior for stochastic approximation and diffusions with slowly decreasing noise effects: Global minimization via Monte Carlo. SIAM J. Appl. Math., 47, 1987, p. 169-185. | Zbl | MR

[17] H.J. Kushner et D.S. Clark, Stochastic approximation for constrained and unconstrained systems. Applied math. science series, vol. 26, Springer, 1978. | Zbl | MR

[18] T.Z. Lai et C.Z. Wei, A note on martingale difference sequences satisfying the local Marcinkiewicz-Zygmund condition. Bull. of the institute of mathematics, Academia Sinica, vol. 11, 1983, p. 1-13. | Zbl | MR

[19] V.A. Lazarev, Convergence of stochastic-approximation procedures in the case of a regression equation with several roots. Problems of Information Transmission, vol. 28, 1992, p. 66-78; en russe, Problemy Peredachi Informatsii, vol. 28, 1992, p. 75-88. | Zbl | MR

[20] P. Lévy, Sur les séries dont les termes sont des variables éventuellement indépendantes. Studia math., vol. 3, 1931, p. 119-155. | Zbl | JFM

[21] L. Ljung, Analysis of recursive stochastic algorithms. IEEE Trans. Automatic Control, vol. 22, 1977, p. 551-575. | Zbl | MR

[22] L. Ljung, G. Pflug et H. Walk, Stochastic approximation of random systems. Birkhäuser, 1992. | Zbl | MR

[23] L. Ljung et T. Soderström, Theory and practice of recursive identification, MIT Press, 1983. | Zbl | MR

[24] S.P. Meyn et R.L. Tweedie, Markov chains and stochastic stability. Springer, 1993. | Zbl | MR

[25] M.B. Nevel'Son et R.Z. Has'Minskii, Stochastic approximation and recursive estimations. Nauka, Moscou, 1972 - Translation of math. monographs, vol. 47, American Mathematical Society, 1973. | Zbl | MR

[26] E. Nummelin, General irreductible Markov chains and nonnegative operators. Cambridge university press, 1984. | Zbl | MR

[27] E. Oja, Principal networks, principal components, and linear neural networks. Neural networks, vol. 5, 1992, p. 927-935.

[28] E. Oja et J. Karhunen, On stochastic approximation of the eigenvalues of the expectation of a random matrix. J. of math. analysis and appl., vol. 106, 1985, p. 69-84. | Zbl | MR

[29] H. Poincaré, Mémoire sur les courbes définies par une équation différentielle (IV). J. Math. Pures Appl., vol. 4, 1886, p. 151-217. | JFM | Numdam

[30] H. Robbins et S. Monro, A stochastic approximation method. Ann. math. stat., vol. 22, 1951, p. 400-407. | Zbl | MR

[31] L. Younes, Estimation and annealing gor Gibbsian fields. Ann. Inst. Henri Poincaré, vol. 24, 1988, p. 269-294. | Zbl | MR | Numdam

[32] L. Younes, Parametric inference of imperfectly observed Gibbsian fields. Probability theory and related fields, vol. 82, 1989, p. 625-645. | Zbl | MR

[33] C.Z. Wei, Martingale transforms with non-atomic limits and stochastic approximation. Probability theory and related fieds, vol. 95, 1993, p. 103-114. | Zbl | MR

R. Pemantle, Non convergence to unstable points in urn models and stochastic approximations. Ann. of Probability, vol. 18, 1990, p. 698-712. | Zbl | MR