Skip to main content
SpringerLink
Log in

On composing an algorithm portfolio

  • Regular Research Paper
  • Published:
Memetic Computing Aims and scope Submit manuscript

Abstract

Evolutionary algorithms are versatile optimization techniques inspired by processes in nature. So far, a wide variety of algorithms have been suggested. However, there is relatively little effort on studying how individual algorithms can work together in a portfolio to achieve a synergy. In this paper, we propose a general methodology to automatically compose a good portfolio from a set of selected evolutionary algorithms. As a single algorithm is a degenerate portfolio, our method also provides an answer to when a portfolio of two or more algorithms are beneficial. Our method has the nice property of being parameter-less; it does not introduce extra parameters. Hence there is no need for parameter control. To illustrate our ideas, we show how a portfolio of five state of the art evolutionary algorithms is automatically constructed using the test functions from the special session on real-parameter optimization of Congress on Evolutionary Computation 2005. It is found that the resulting portfolio obtains the best average ranking. The applicability and limitations of the paradigm of using a benchmarking suite to access evolutionary algorithms are also examined. Though this paper has used evolutionary algorithms only to compose an algorithm portfolio, the idea is generic and is applicable to portfolios with non-evolutionary algorithms as well.

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

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. Engelbrecht AP (2007) Computational intelligence, an introduction, 2nd edn. Wiley, New Jersey

    Book  Google Scholar 

  2. Brownlee J (2011) Clever algorithms: nature-inspired programming pecipes, 1st edn. Lulu Enterprises, Australia

    Google Scholar 

  3. Lam AYS, Li VOK (2012) Real-coded chemical reaction optimization. IEEE Trans Evol Comput 16:339–353

    Article  Google Scholar 

  4. Taha HA (1997) Operations research: an introduction, 6th edn. Prentice Hall, New Jersey

    MATH  Google Scholar 

  5. Nguyen QH, Ong Y-S, Lim MH (2009) A probabilistic memetic framework. IEEE Trans Evol Comput 13:604–623

    Article  Google Scholar 

  6. Wolpert DH, Macready WG (1997) No free lunch theorems for optimization. IEEE Trans Evol Comput 1:67–82

    Article  Google Scholar 

  7. Koehler GJ (2007) Conditions that obviate the no-free-lunch theorems for optimization. INFORMS J Comput 19:273–279

    Article  MathSciNet  MATH  Google Scholar 

  8. Vrugt JA, Robinson BA (2007) Improved evolutionary optimization from genetically adaptive multimethod search. Proc Natl Acad Sci 104:708–711

    Article  Google Scholar 

  9. Krasnogor N, Smith J (2005) A tutorial for competent memetic algorithms: model, taxonomy and design issues. IEEE Transs Evol Comput 9:474–488

    Article  Google Scholar 

  10. Munoz MA, Kirley M, Halgamuge SK (2013) The algorithm selection problem on the continuous optimization domain. In: Moewes C, Nurnberger A (eds) Computational intelligence in intelligent data analysis, SCI 445. Springer, Heidelberg, pp 75–89

    Chapter  Google Scholar 

  11. Gomes CP, Selman B (2001) Algorithm portfolios. Artif Intell 126:43–62

    Article  MathSciNet  Google Scholar 

  12. Dietterich TG (2000) Ensemble methods in machine learning, vol 1857., Lecture notes in computer scienceSpringer, Heidelberg

    Google Scholar 

  13. Burke EK, Gendreau M, Hyde M, Kendall G, Ochoa G, Ozcan E, Qu R (2013) Hyper-heuristics: a survey of the state of the art. J Oper Res Soc 64:1695–1724

    Article  Google Scholar 

  14. Qin AK, Huang VL, Suganthan PN (2009) Differential evolution algorithm with strategy adaptation for global numerical optimization. IEEE Trans Evol Comput 13:398–417

    Article  Google Scholar 

  15. Mallipeddi R, Suganthan PN, Pan QK, Tasgetiren MF (2011) Differential evolution algorithm with ensemble of parameters and mutation strategies. Appl Soft Comput 11:1679–1696

    Article  Google Scholar 

  16. Du W, Li B (2008) Multi-strategy ensemble particle swarm optimization for dynamic optimization. Inf Sci 178:3096–3109

    Article  MATH  Google Scholar 

  17. Vrugt JA, Robinson BA, Hyman JM (2009) Self-adaptive multimethod search for global optimization in real-parameter spaces. IEEE Trans Evol Comput 13:243–259

    Article  Google Scholar 

  18. Peng F, Tang K, Chen G, Yao X (2010) Population-based algorithm portfolios for numerical optimization. IEEE Trans Evol Comput 14:782–800

    Article  Google Scholar 

  19. Wang Y, Li B (2010) Multi-strategy ensemble evolutionary algorithm for dynamic multi-objective optimization. Memetic Comput 2:3–24

    Article  MATH  Google Scholar 

  20. Grobler J, Engelbrecht AP, Kendall G, Yadavalli VSS (2013) Multi-method algorithms: investigating the entity-to-algorithm allocation problem. In: Proc. IEEE congress on evolutionary computation. IEEE Press, New York, pp 570–577

  21. Hadka D, Reed P (2013) Borg: an auto-adaptive many-objective evolutionary computing framework. Evol Comput 21:231–259

    Article  Google Scholar 

  22. Li K, Fialho A, Kwong S, Zhang Q (2014) Adaptive operator selection with bandits for a multiobjective evolutionary algorithm based on decomposition. IEEE Trans Evol Comput 18:114–130

    Article  Google Scholar 

  23. Hansen N (2011) The CMA evolutionary strategy: a tutorial. Tech. Rep., http://www.lri.fr/~hansen/cmatutorial.pdf. Accessed 23 Feb 2015

  24. Black-Box Optimization Benchmarking (BBOB) (2013) http://coco.gforge.inria.fr/doku.php?id=bbob-2013. Accessed 23 Feb 2015

  25. Das S, Suganthan PN (2011) Differential evolution: a survey of the state-of-the-art. IEEE Trans Evol Comput 15:4–31

    Article  Google Scholar 

  26. Wang Y, Cai Z, Zhang Q (2011) Differential evolution with composite trial vector generation strategies and control parameters. IEEE Trans Evol Comput 15:55–66

    Article  Google Scholar 

  27. Particle Swarm Central (2015) http://www.particleswarm.info/. Accessed 23 Feb 2015

  28. Karaboga D, Basturk B (2007) A powerful and efficient algorithm for numerical function optimization: artificial bee colony (ABC) algorithm. J Glob Optim 39:459–471

    Article  MathSciNet  MATH  Google Scholar 

  29. Yuen SY, Chow CK, Zhang X (2013) Which algorithm should I choose at any point of the search: an evolutionary portfolio approach. In: Proc. of the 14th Int. Conf. on Genetic and Evol. Comput. Conf (GECCO). ACM, New York, pp 567–574

  30. Lobo FJ, Lima CF, Michalewicz Z (2007) Parameter setting in evolutionary algorithms. Springer, Berlin

  31. Yuen SY, Zhang X (2013) On Composing an (evolutionary) algorithm portfolio. In: Proc. of the 14th Int. Conf. on Genetic and Evol. Comput. Conf (GECCO). ACM, New York, pp 83–84

  32. Suganthan PN, Hansen N, Liang JJ, Deb K, Chen YP, Auger A, Tiwari S (2005) Problem Definitions and Evaluation Criteria for the CEC 2005 Special Session on Real-parameter Optimization”, Nanyang Technol. Univ. and IIT Kanpur, Singapore and Kanpur, India, Tech. Rep. 2005005

  33. Garcia S, Molina D, Lozano M, Herrera F (2009) A study on the use of non-parametric tests for analyzing the evolutionary algorithms’ behaviour a case study on the CEC’2005 special session on real parameter optimization. J Heuristics 15:617–644

    Article  MATH  Google Scholar 

  34. Mann HB, Whitney DR (1947) On a test of whether one of two random variables is stochastically larger than the other. Ann Math Stat 18:50–60

    Article  MathSciNet  MATH  Google Scholar 

  35. Dowold K, Aderhold A, Scheidler A, Middendorf M (2011) Performance evaluation of artificial bee colony optimization and new selection schemes. Memetic Comput 3:149–162

    Article  Google Scholar 

  36. Sörensen K (2015) Metaheuristics—the metaphor exposed. Int Trans Oper Res 22:3–18

  37. Liang JJ, Qu B-Y, Suganthan PN, Hernández-Díaz AG (2013) Problem definitions and evaluation criteria for the CEC 2013 special session and competition on real-parameter optimization. Technical report, Nanyang Technological University, Singapore

  38. Tang K, Peng F, Chen G, Yao X (2014) Population-based algorithm portfolios with automated constituent algorithms selection. Inf Sci 279:94–104

    Article  Google Scholar 

Download references

Acknowledgments

The work described in this paper was supported by a grant from the Research Grants Council of the Hong Kong Special Administrative Region, China [Project No. CityU 125313]. We thank Yang Lou and Yau King Lam for proofreading the manuscript.

Author information

Authors and Affiliations

  1. Department of Electronic Engineering, City University of Hong Kong, Hong Kong, China

    Shiu Yin Yuen

  2. College of Electronic and Communication Engineering, Tianjin Normal University, Tianjin, China

    Xin Zhang

Authors
  1. Shiu Yin Yuen
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Xin Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Shiu Yin Yuen.

Appendices

Appendix 1

See Appendix Tables 8, 9, 10, 11 and 12.

Table 8 The mean and standard deviation (SD) of function error values attained by each algorithm for \(f_{1}-f_{5}\) with \(D=30\)
Full size table
Table 9 The mean and standard deviation (SD) of function error values attained by each algorithm for \(f_{6}-f_{10}\) with \(D=30\)
Full size table
Table 10 The mean and standard deviation (SD) of function error values attained by each algorithm for \(f_{11}-f_{15}\) with \(D = 30\)
Full size table
Table 11 The mean and standard deviation (SD) of function error values attained by each algorithm for \(f_{16}-f_{20}\) with \(D= 30\)
Full size table
Table 12 The mean and standard deviation (SD) of function error values attained by each algorithm for \(f_{21}-f_{25}\) with \(D= 30\)
Full size table

Appendix 2

See Appendix Table 13.

Table 13 The mean and standard deviation (SD) of function error values attained by MultiEA1 for \(f_{1}-f_{25}\) with \(D= 30\)
Full size table

Appendix 3

See Appendix Table 14.

Table 14 The mean and standard deviation (SD) of function error values attained by MultiEA2 for \(f_{1}-f_{25}\) with \(D= 30\)
Full size table

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Yuen, S.Y., Zhang, X. On composing an algorithm portfolio. Memetic Comp. 7, 203–214 (2015). https://doi.org/10.1007/s12293-015-0159-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s12293-015-0159-9

Keywords

Search

Navigation