Abstract
In this paper, we propose a method of determining the initial temperature for continuous fast simulated annealing from the perspective of state variation. While the conventional method utilizes fitness variation, the proposed method additionally considers genotype variation. The proposed scheme is based on the fact that the annealing temperature, which includes the initial temperature, not only appears in the acceptance probability but serves as the scale parameter of a state generating probability distribution. We theoretically derive an expression for the probability of generating states to cover the state space in conjunction with the convergence property of the fast simulated annealing. We then numerically solve the expression to determine the initial temperature. We empirically show that the proposed method outperforms the conventional one in optimizing various benchmarking functions.
Similar content being viewed by others
References
Kirkpatrick, S., Gelatt, C., Vecchi, M.: Optimization by simulated annealing. Science 220, 671–680 (1983)
Černý, V.: A thermodynamical approach to the traveling salesman problem: an efficient simulation algorithm. J. Optim. Theory Appl. 45, 41–51 (1985)
Johnson, D., Aragon, C., Mcgeoch, L., Schevon, C.: Optimization simulated annealing: an experimental evaluation; Part I, graph partitioning. Oper. Res. 37, 865–891 (1989)
Geman, S., Geman, D.: Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images. IEEE Trans. Pattern Anal. Mach. Intell. PAMI-6, 721–741 (1984)
Szu, H., Hartley, R.: Fast simulated annealing. Phys. Lett. A 122, 157–162 (1987)
Nascimento, V., Carvalho, V., de Castilho, C., Costa, B., Soares, E.: The fast simulated annealing algorithm applied to the search problem in LEED. Surf. Sci. 487, 15–27 (2001)
Szu, H.: Automatic fault recognition by image correlation neural network techniques. IEEE Trans. Ind. Electron. 40, 197–208 (1993)
Mageras, G., Mohan, R.: Application of fast simulated annealing to optimization of conformal radiation treatments. Med. Phys. 20, 639–647 (1993)
Wang, Z., Hanson, J.: Codebook optimization by Cauchy annealing. In: The Proceedings of Global Telecommunications Conference, vol. 3, pp. 1974–1978 (1993)
Ingber, L.: Very fast simulated re-annealing. Math. Comput. Model. 12, 967–973 (1989)
Tsallis, C., Stariolo, D.: Generalized simulated annealing. Physica A 233, 395–406 (1996)
Metropolis, N., et al.: Equations of state calculations by fast computing machines. J. Chem. Phys. 21, 1087–1092 (1953)
Suman, B., Kumar, P.: A survey of simulated annealing as a tool for single and multiobjective optimization. J. Oper. Res. Soc. 57, 1143–1160 (2006)
Johnson, D., Aragon, C., Mcgeoch, L., Schevon, C.: Optimization simulated annealing: an experimental evaluation; Part II, graph coloring and number partitioning. Oper. Res. 39, 378–406 (1991)
Chen, H., Flann, N., Watson, D.: Parallel genetic simulated annealing: a massively parallel SIMD algorithm. IEEE Trans. Parallel Distrib. Syst. 9, 126–136 (1998)
Pao, D., Lam, S., Fong, A.: Parallel implementation of simulated annealing using transaction processing. IEE Proc., Comput. Digit. Tech. 146, 107–113 (1999)
Thompson, D., Bilbro, G.: Sample-sort simulated annealing. IEEE Trans. Syst. Man Cybern., Part B, Cybern. 35, 625–632 (2005)
Xinchao, Z.: Simulated annealing algorithm with adaptive neighborhood. Appl. Soft Comput. 11(2), 1827–1836 (2011)
Ben-Ameur, W.: Computing the initial temperature of simulated annealing. Comput. Optim. Appl. 29, 369–385 (2004)
Oliveira, H.A. Jr., Petraglia, A., Ingber, L., Machado, M.A.S., Petraglia, M.R.: Stochastic Global Optimization and Its Applications with Fuzzy Adaptive Simulated Annealing. Springer, New York (2012)
Kincard, D., Cheney, W.: Numerical Analysis: Mathematics of Scientific Computing. Brooks/Coles, Pacific Grove (2002). Chap. 3
Hedar, A., Fukushima, M.: Tabu search directed by direct search methods for nonlinear global optimization. Eur. J. Oper. Res. 170, 329–349 (2006)
Schaffer, J., Caruana, R., Eshelman, L., Das, R.: A study of control parameters affecting online performance of genetic algorithms for function optimization. In: Schaffer, J. (ed.) Proc. of the 3rd Int’l Conf. on GAs, pp. 51–60. Morgan Kauffman, San Francisco (1989)
Lee, C.-Y., Yao, X.: Evolutionary programming using mutations based on the Lévy probability distribution. IEEE Trans. Evol. Comput. 8, 1–13 (2004)
Chao, L.: Statistics: Methods and Analyses. McGraw-Hill, New York (1966)
Acknowledgement
This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2012-0007627).
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
In the Appendix, we derive Eq. (9) from
Suppose that a state x has the range R=U−L, where x∈[L, U], and we divide the range into N intervals with widths of 2α so that N=R/(2α). For convenience, rank the centers of N intervals, \(x_{j}^{c}\) (j=1,2,…,N), in ascending order in their position. That is, we assume, without a loss of generality, that \(x_{i}^{c}\)’s are ordered in such a way that \(x_{i}^{c} > x_{j}^{c}\) when i>j. With this convention, we can express the center of j-th interval as
where j=1,2,…,N, so that we have
Denote \(x_{j}^{\ast}\) (j=1,2,…,N) as a sample state generated by a Cauchy distribution of the location parameter \(x_{j}^{c}\). Using the probability density function of the Cauchy distribution of Eq. (6), we can express
Thus, Eq. (30) becomes
To compute the integrals in Eq. (35), consider the following transformation:
With the transformation, we have
Thus, the first integral on the right hand side of Eq. (35) can be rewritten as
Similarly, the second integral can be rewritten as
With Eqs. (38) and (39), we can rewrite Eq. (30) as
where
and we have used
Rights and permissions
About this article
Cite this article
Lee, CY., Lee, D. Determination of initial temperature in fast simulated annealing. Comput Optim Appl 58, 503–522 (2014). https://doi.org/10.1007/s10589-013-9631-y
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10589-013-9631-y