World Scientific
  • Search
    Quick Search in Journals
    Access type:
    Quick Search anywhere
    Access type:
    Advanced Search
  • My Cart
Skip main navigation
Close Drawer MenuOpen Drawer Menu

Cookies Notification

We use cookies on this site to enhance your user experience. By continuing to browse the site, you consent to the use of our cookies. Learn More
×
Our website is made possible by displaying certain online content using javascript.
In order to view the full content, please disable your ad blocker or whitelist our website www.worldscientific.com.

System Upgrade on Tue, Oct 25th, 2022 at 2am (EDT)

Existing users will be able to log into the site and access content. However, E-commerce and registration of new users may not be available for up to 12 hours.
For online purchase, please visit us again. Contact us at [email protected] for any enquiries.
CORRESPONDENCE: OptimizationNo Access

A UNIFORM ENHANCEMENT APPROACH FOR OPTIMIZATION ALGORITHMS: SMOOTHING FUNCTION METHOD

    https://doi.org/10.1142/S0218001410008287Cited by:3
    Previous
    Next

    Abstract

    In this paper, we propose a uniform enhancement approach called smoothing function method, which can cooperate any optimization algorithm and improve its performance. The method has two phases. In the first phase, a smoothing function is constructed by using a properly truncated Fourier series. It can preserve the overall shape of the original objective function but eliminate many of its local optimal points, thus it can well approach the objective function. Then, the optimal solution of the smoothing function is searched by an optimization algorithm (e.g. traditional algorithm or evolutionary algorithm) so that the search becomes much easier. In the second phase, we switch to optimize the original function for some iterations by using the best solution(s) obtained in phase 1 as an initial point (population). Thereafter, the smoothing function is updated in order to approximate the original function more accurately.

    These two phases are repeated until the best solutions obtained in several successively second phases cannot be improved obviously. In this manner, any optimization algorithm will become much easier in searching optimal solution. Finally, we use the proposed approach to enhance two typical optimization algorithms: Powell direct algorithm and a simple genetic algorithm. The simulation results on ten challenging benchmarks indicate the proposed approach can effectively improve the performance of these two algorithms.

    References

    • P. Amara, D. Hsu and J. E. Straub, Comput. Phys. 97, 6715 (1993). Google Scholar
    • I. Andricioaei and J. E. Straub, Comput. Phys. 10, 449 (1996), DOI: 10.1063/1.168582. Crossref, Google Scholar
    • M. S.   Bazaraa , Nonlinear Programming: Theory and Algorithms ( Wiley , New York , 1993 ) . Google Scholar
    • M.   Cartwright , Fourier Methods for Mathematicians, Scientists and Engineers ( Ellis Horwood Limited , Chichester, England , 1990 ) . Google Scholar
    • M. Clerc and J. Kennedy, IEEE Trans. Evolut. Comput. 6(1), 58 (2002), DOI: 10.1109/4235.985692. Crossref, ISI, Google Scholar
    • S. C. Esquivel, H. A. Leiva and R. H. Gallard, Evolutionary Computation, eds. A. E. Eiben and Z. Michalewicz (IOS Press, Amsterdam, Netherlands, 1999) pp. 17–33. Google Scholar
    • F. T.   Fang and Y.   Wang , Number-Theoretic Methods in Statistics ( Chapman & Hall , London, UK , 1994 ) . Crossref, Google Scholar
    • R.   Fletcher , Practical Methods of Optimization ( Wiley , Chichester, England , 1987 ) . Google Scholar
    • G. W. Greenwood, G. B. Fogel and M. Ciobanu, Emphasizing extinction in evolutionary programming, Proc. 1999 Congress on Evolutionary Computation1 (1999) pp. 666–671. Google Scholar
    • J. R.   Hanna and J. H.   Rowland , Fourier Series, Transforms, and Boundary Value Problems ( Wiley , New York , 1990 ) . Google Scholar
    • J.-P. Kahane et al. , Fourier Series and Wavelets ( Gordon and Breach Publishers , Luxembourg , 1995 ) . Google Scholar
    • C. T.   Kelley , Iterative Methods for Optimization ( SIAM , Philadelphia , 1999 ) . Crossref, Google Scholar
    • Y. W. Leung and Y. P. Wang, IEEE Trans. Syst. Man Cybern. Part C 30(3), 293 (2000). ISI, Google Scholar
    • J. Ma, D. Hsu and J. E. Straub, J. Chem. Phys. 99, 4024 (1993), DOI: 10.1063/1.466098. Crossref, ISI, Google Scholar
    • Z.   Michalewicz , Genetic Algorithms + Data Structures = Evolution Programs , 3rd edn. ( Springer-Verlag , Berlin , 1996 ) . Crossref, Google Scholar
    • L. Piela, J. Kostrowicki and H. A. Scheraga, J. Phys. Chem. 93, 3339 (1989), DOI: 10.1021/j100345a090. Crossref, ISI, Google Scholar
    • J. Pillardy, K. A. Olszewski and L. Piela, J. Phys. Chem. 99, 4337 (1992). Google Scholar
    • J. Pillardy and L. Piela, J. Comp. Chem. 19, 2040 (1998). Google Scholar
    • J. Renders and H. Bersini, Hybridizing genetic algorithms with hill-climbing methods for global optimization: Two possible ways, Proc. First IEEE Int. Conf. Evolutionary Computation (IEEE Press, Orlando, USA, 1994) pp. 312–317. Google Scholar
    • R. L. Somorjai, J. Phys. Chem. 95, 4141 (1991), DOI: 10.1021/j100163a045. Crossref, ISI, Google Scholar
    • J. E. Straub, J. Ma and P. Amara, J. Chem. Phys. 103, 1574 (1995), DOI: 10.1063/1.469779. Crossref, ISI, Google Scholar
    • M. A. Stybinski and T. S. Tang, Neural Networks 3, 467 (1990). Crossref, ISI, Google Scholar
    • M. Sylvain and R. L. Somorjai, J. Phys. Chem. 95, 4147 (1991), DOI: 10.1021/j100163a046. Crossref, ISI, Google Scholar
    • H. Verscheldeet al., J. Chem. Phys. 106, 1556 (1997), DOI: 10.1063/1.473277. Crossref, ISI, Google Scholar
    • R. J. Wawaket al., Proc. Nat. Acad. Sci. USA 93, 1743 (1996), DOI: 10.1073/pnas.93.5.1743. Crossref, ISI, Google Scholar
    • X. Yao and Y. Liu, Evolutionary Programming VI, eds. P. J. Angelineet al. (Springer-Verlag, Berlin, 1997) pp. 151–161. Crossref, Google Scholar
    • X. Yao, Y. Liu and G. Lin, IEEE Trans. Evolut. Comput. 3(2), 82 (1999). ISI, Google Scholar
    • J. Yen and B. Lee, A simplex genetic algorithm hybrid, Proc. 1997 IEEE Int. Conf. Evolutionary Computation (IEEE Press, Indiana, USA, 1997) pp. 175–180. Google Scholar