Abstract
Motivated by the recent developments in digital diffusion networks, this work is devoted to the rates of convergence issue for a class of global optimization algorithms. By means of weak convergence methods, we show that a sequence of suitably scaled estimation errors converges weakly to a diffusion process (a solution of a stochastic differential equation). The scaling together with the stationary covariance of the limit diffusion process gives the desired rates of convergence. Application examples are also provided for some image estimation problems.
Similar content being viewed by others
References
X. Cai, P. Kelly, and W. B. Gong, Digital diffusion network for image segmentation, Proc. IEEE Internat. Conf. Image Processing, 1995.
T. S. Chiang, C.R. Hwang, and S.J. Sheu, Diffusion for global optimization in ℝn, SIAM J. Control Optim. 25 (1987), 737-752.
M. H. Dowell, A Parallel Implementation of a Digital Diffusion Network for Image Segmentation, M.S. Thesis, Department of Electrical and Computer Engineering, University of Massachusetts, Amherst, 1999.
S. N. Ethier and T. G. Kurtz, Markov Processes, Characterization and Convergence, Wiley, New York, 1986.
S. B. Gelfand and S. K. Mitter, Recursive stochastic algorithms for global optimization in ℝd, SIAM J. Control Optim. 29 (1991), 999-1018.
J. Hokland and P. Kelly, Markov models of specular and diffuse scattering in restoration of medical ultrasound images, IEEE Trans. Ultrasonics, Ferroelectrics and Frequency Control 43 (1996), 660-669.
G. Kesidis, Analog optimization with Wong's stochastic neural network, IEEE Trans. Neural Net. 6 (1995), 258-260.
S. Kirkpatrick, C. D. Gelatt, and M. P. Vecchi, Optimization by simulated annealing, Science 220 (1983), 671-680.
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), 169-185.
H. J. Kushner, Approximation and Weak Convergence Methods for Random Processes, with applications to Stochastic Systems Theory, MIT Press, Cambridge, MA, 1984.
H. J. Kushner and G. Yin, Stochastic Approximation Algorithms and Applications, Springer-Verlag, New York, 1997.
P. L'Ecuyer and G. Yin, Budget-dependent rate of stochastic approximation, SIAM J. Optim. 8 (1998), 217-247.
B. Manjunath, T. Simchony, and R. Chellappa, Stochastic and deterministic networks for texture segmentation, IEEE Trans. ASSP 38 1990.
M. Metropolis, A. W. Rosenbluth, M. N. Rosenbluth, A.H. Teller, and E. Teller, Equations of state calculations by fast computing machines, J. Chem. Phys. 21 (1953), 1087-1091.
E. Wong, Stochastic neural networks, Algorithmica 6 (1991), 466-478.
G. Yin, Rates of convergence for a class of global stochastic optimization algorithms, SIAM J. Optim., 10 (1999), 99-120.
G. Yin, P.A. Kelly, and M.H. Dowell, Approximation of an analog diffusion network with applications to image estimation, J. Optim. Theory Appl. 107 (2000), 391-414.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Yin, G., Kelly, P. Convergence Rates of Digital Diffusion Network Algorithms for Global Optimization with Applications to Image Estimation. Journal of Global Optimization 23, 329–358 (2002). https://doi.org/10.1023/A:1016539015160
Issue Date:
DOI: https://doi.org/10.1023/A:1016539015160