Springer Nature is making SARS-CoV-2 and COVID-19 research free. View research | View latest news | Sign up for updates

Annealing evolutionary stochastic approximation Monte Carlo for global optimization

Abstract

In this paper, we propose a new algorithm, the so-called annealing evolutionary stochastic approximation Monte Carlo (AESAMC) algorithm as a general optimization technique, and study its convergence. AESAMC possesses a self-adjusting mechanism, whose target distribution can be adapted at each iteration according to the current samples. Thus, AESAMC falls into the class of adaptive Monte Carlo methods. This mechanism also makes AESAMC less trapped by local energy minima than nonadaptive MCMC algorithms. Under mild conditions, we show that AESAMC can converge weakly toward a neighboring set of global minima in the space of energy. AESAMC is tested on multiple optimization problems. The numerical results indicate that AESAMC can potentially outperform simulated annealing, the genetic algorithm, annealing stochastic approximation Monte Carlo, and some other metaheuristics in function optimization.

This is a preview of subscription content, log in to check access.

References

  1. Alcock, J., Burrage, K.: A genetic algorithm for parameters of stochastic ordinary differential equations. Comput. Stat. Data Anal. 47, 255–275 (2004)

  2. Ali, M.M., Storey, C.: Modified controlled random search algorithms. Int. J. Comput. Math. 54, 229–235 (1994)

  3. Ali, M.M., Khompatraporn, C., Zabinsky, Z.B.: A numerical evaluation of several stochastic algorithms on selected continuous global optimization test problems. J. Glob. Optim. 31, 635–672 (2005)

  4. Andrieu, C., Robert, C.P.: Controlled MCMC for optimal sampling. Technical Report 0125, Cahiers du Cérémade (2001)

  5. Andrieu, C., Moulines, É., Priouret, P.: Stability of stochastic approximation under verifiable conditions. SIAM J. Control Optim. 44, 283–312 (2005)

  6. Angelis, L., Bora-Senta, E., Moyssiadis, C.: Optimal exact experimental designs with correlated errors through a simulated annealing. Comput. Stat. Data Anal. 37, 275–296 (2001)

  7. Atchadé, Y., Liu, J.S.: The Wang-Landau algorithm for Monte Carlo computation on general state-spaces. Technical Report, University of Michigan (2009)

  8. Baragona, R., Battaglia, F., Calzini, C.: Genetic algorithms for the identification of additive and innovation outliers in time series. Comput. Stat. Data Anal. 37, 1–12 (2001)

  9. Billingsley, P.: Probability and Measure, 2nd edn. Wiley, New York (1986)

  10. Cappé, O., Guillin, A., Marin, J.M., Robert, C.P.: Population Monte Carlo. J. Comput. Graph. Stat. 13, 907–929 (2004)

  11. Dorea, C.C.Y.: Stopping rules for a random optimization method. SIAM J. Control Optim. 28, 841–850 (1990)

  12. Dorsey, R.E., Mayer, W.J.: Genetic algorithms for estimation problems with multiple optima, non-differentiability, and other irregular features. J. Bus. Econ. Stat. 13, 53–66 (1995)

  13. Duczmal, L., Assuncão, R.: A simulated annealing strategy for the detection of arbitrarily shaped spatial clusters. Comput. Stat. Data Anal. 45, 269–286 (2004)

  14. Duczmal, L., Cancado, A.L.F., Takahashi, R.H.C., Bessegato, L.F.: A genetic algorithm for irregularly shaped spatial scan statistics. Comput. Stat. Data Anal. 52, 43–52 (2007)

  15. Eberhart, R.C., Kennedy, J.: A new optimizer using particle swarm theory. In: Proc. 6th Int. Symp. Micromachine Human Sci., Nagoya, Japan, pp. 39–43 (1995)

  16. Ferri, M., Piccioni, M.: Optimal selection of statistical units: an approach via simulated annealing. Comput. Stat. Data Anal. 13, 47–61 (1992)

  17. Geman, S., Geman, D.: Stochastic relaxation, Gibbs distribution and the Bayesian restoration of images. IEEE Trans. Pattern Anal. Mach. Intell. 6, 721–741 (1984)

  18. Gilks, W.R., Roberts, G.O., George, E.I.: Adaptive direction sampling. Statistician 43, 179–189 (1994)

  19. Gilks, W.R., Roberts, G.O., Sahu, S.K.: Adaptive Markov chain Monte Carlo through regeneration. J. Am. Stat. Assoc. 93, 1045–1054 (1998)

  20. Glover, F., Laguna, M.: Tabu Search. Kluwer Academic, Dordrecht (1997)

  21. Goldberg, D.E.: Genetic Algorithms in Search, Optimization, & Machine Learning. Addison-Wesley, Reading (1989)

  22. Goswami, G.R., Liu, J.S.: On learning strategies for evolutionary Monte Carlo. Stat. Comput. 17, 23–38 (2007)

  23. Harrio, H., Saksman, E., Tamminen, J.: An adaptive Metropolis algorithm. Bernoulli 7, 223–242 (2001)

  24. Hart, W.E.: Sequential stopping rules for random optimization methods with application to multistart local search. SIAM J. Optim. 9, 270–290 (1998)

  25. Hastings, W.K.: Monte Carlo sampling methods using Markov chains and their applications. Biometrika 57, 97–109 (1970)

  26. Hedar, A.R., Fukushima, M.: Tabu search directed by direct search methods for nonlinear global optimization. Eur. J. Oper. Res. 170, 329–349 (2006)

  27. Hesselbo, B., Stinchcomble, R.B.: Monte Carlo simulation and global optimization without parameters. Phys. Rev. Lett. 74, 2151–2155 (1995)

  28. Hirsch, M.J., Pardalos, P.M., Resende, M.G.C.: Speeding up continuous GRASP. Eur. J. Oper. Res. (2006, submitted)

  29. Hirsch, M.J., Meneses, C.N., Pardalos, P.M., Resende, M.G.C.: Global optimization by continuous GRASP. Optim. Lett. 1, 201–212 (2007)

  30. Hoffman, D.L., Schmidt, P.: Testing the restrictions implied by the rational expectations hypothesis. J. Econom. 15, 265–287 (1981)

  31. Holland, J.H.: Adaptation in Natural and Artificial Systems. University of Michigan Press, Ann Arbor (1975)

  32. Jasra, A., Stephens, D.A., Holmes, C.C.: On population-based simulation for static inference. Stat. Comput. 17, 263–279 (2007)

  33. Kirkpatrick, S., Gelatt, C.D., Vecchi, M.P.: Optimization by simulated annealing. Science 220, 671–680 (1983)

  34. Laguna, M., Martí, R.: Experimental testing of advanced scatter search designs for global optimization of multimodal functions. J. Glob. Optim. 33, 235–255 (2005)

  35. Liang, F.: Dynamically weighted importance sampling in Monte Carlo computation. J. Am. Stat. Assoc. 97, 807–821 (2002)

  36. Liang, F.: Annealing stochastic approximation Monte Carlo for neural network training. Mach. Learn. 68, 201–233 (2007)

  37. Liang, F., Wong, W.H.: Evolutionary Monte Carlo sampling: applications to C p model sampling and change-point problem. Stat. Sin. 10, 317–342 (2000)

  38. Liang, F., Wong, W.H.: Real parameter evolutionary sampling with applications in Bayesian mixture models. J. Am. Stat. Assoc. 96, 653–666 (2001)

  39. Liang, F., Liu, C., Carroll, R.J.: Stochastic approximation in Monte Carlo computation. J. Am. Stat. Assoc. 102, 305–320 (2007)

  40. Liang, J.J., Qing, A.K., Suganthan, P.N., Baskar, S.: Comprehensive learning particle swarm optimizer for global optimization of multimodal functions. IEEE Trans. Evol. Comput. 10, 281–295 (2006)

  41. Liu, J.S.: Monte Carlo Strategies in Scientific Computing. Springer, New York (2001)

  42. Mengersen, K.L., Tweedie, R.L.: Rates of convergence of the Hastings and Metropolis algorithms. Ann. Stat. 24, 101–121 (1996)

  43. Metropolis, N., Rosenbluth, A.W., Rosenbluth, M.N., Teller, A.H., Teller, E.: Equation of state calculations by fast computing machines. J. Chem. Phys. 21, 1087–1091 (1953)

  44. Meyn, S.P., Tweedie, R.L.: Markov Chains and Stochastic Stability. Springer, London (1993)

  45. Michalewicz, Z., Nazhiyath, G.: Genocop III: A co-evolutionary algorithm for numerical optimization problems with nonlinear constraints. In: Proceedings of the Second IEEE ICEC, Perth, Australia (1995)

  46. Müller, P.: A generic approach to posterior integration and Gibbs sampling. Technical Report, Purdue University, West Lafayette, Indiana (1991)

  47. Price, W.L.: Global optimization by controlled random search. J. Optim. Theory Appl. 40, 333–348 (1983)

  48. Robert, C.P., Casella, G.: Monte Carlo Statistical Methods, 2nd edn. Springer, New York (2004)

  49. Salomon, R.: Reevaluating genetic algorithm performance under coordinate rotation of benchmark functions. BioSystems 39, 263–278 (1996)

  50. Schmitt, L.M.: Theory of genetic algorithms. Theor. Comput. Sci. 259, 1–61 (2001)

  51. Schmitt, L.M.: Asymptotic convergence of scaled genetic algorithms to global optima. In: Menon, A. (ed.) Frontiers of Evolutionary Computation, pp. 157–200. Kluwer Academic, Dordrecht (2004)

  52. Storn, R., Price, K.: Differential evolution: A simple and efficient heuristic for global optimization over continuous spaces. J. Glob. Optim 11, 341–359 (1997)

  53. Tierney, L.: Introduction to general state-space Markov chain theory. In: Gilks, W.R., Richardson, S., Spiegelhalter, D.J. (eds.) Markov Chain Monte Carlo in Practice, pp. 59–88. Chapman & Hall, London (1996)

  54. Törn, A., Žilinskas, A.: Global Optimization. Lecture Notes in Computer Science, vol. 350. Springer, Berlin (1989)

  55. Winker, P.: Identification of multivariate AR-models by threshold accepting. Comput. Stat. Data Anal. 20, 295–307 (1995)

  56. Wong, W.H., Liang, F.: Dynamic weighting in Monte Carlo and optimization. Proc. Natl. Acad. Sci. USA 94, 14220–14224 (1997)

  57. Wu, B., Chang, C.-L.: Using genetic algorithms to parameters (d,r) estimation for threshold autoregressive models. Comput. Stat. Data Anal. 38, 315–330 (2002)

  58. Zhu, H.T., Liang, F., Gu, M., Peterson, B.: Stochastic approximation algorithms for estimation of spatial mixed models. In: Lee, S.Y. (ed.) Handbook of Computing and Statistics with Applications, vol. 1, pp. 399–421. Elsevier, Amsterdam (2007)

Download references

Author information

Correspondence to Faming Liang.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Liang, F. Annealing evolutionary stochastic approximation Monte Carlo for global optimization. Stat Comput 21, 375–393 (2011). https://doi.org/10.1007/s11222-010-9176-1

Download citation

Keywords

  • Convergence
  • Genetic algorithm
  • Global optimization
  • Simulated annealing
  • Stochastic approximation Monte Carlo