Skip to main content
SpringerLink
Log in

The Impact of Population Size, Number of Children, and Number of Reference Points on the Performance of NSGA-III

  • Conference paper
  • First Online:
Evolutionary Multi-Criterion Optimization (EMO 2017)

Part of the book series: Lecture Notes in Computer Science ((LNTCS,volume 10173))

Included in the following conference series:

Abstract

We investigate the impact of three control parameters (the population size \(\mu \), the number of children \(\lambda \), and the number of reference points H) on the performance of Nondominated Sorting Genetic Algorithm III (NSGA-III). In the past few years, many efficient Multi-Objective Evolutionary Algorithms (MOEAs) for Many-Objective Optimization Problems (MaOPs) have been proposed, but their control parameters have been poorly analyzed. The recently proposed NSGA-III is one of most promising MOEAs for MaOPs. It is widely believed that NSGA-III is almost parameter-less and requires setting only one control parameter (H), and the value of \(\mu \) and \(\lambda \) can be set to \(\mu = \lambda \approx H\) as described in the original NSGA-III paper. However, the experimental results in this paper show that suitable parameter settings of \(\mu \), \(\lambda \), and H values differ from each other as well as their widely used parameter settings. Also, the performance of NSGA-III significantly depends on them. Thus, the usually used parameter settings of NSGA-III (i.e., \(\mu = \lambda \approx H\)) might be unsuitable in many cases, and \(\mu \), \(\lambda \), and H require a particular parameter tuning to realize the best performance of NSGA-III.

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

Access this chapter

Log in via an institution
Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD 39.99
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 54.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Notes

  1. 1.

    The code was downloaded from http://www.cs.bham.ac.uk/~xin/journal_papers.html.

  2. 2.

    Since the IGD indicator [28] used in [6] is unsuitable for comparing nondominated solution sets of different size as pointed out in [10], we did not use it.

  3. 3.

    Since, as far as we know, there is no good diversity indicator for the unbounded archive, we could not measure the diversity of the obtained nondominated solutions.

References

  1. Andersson, M., Bandaru, S., Ng, A., Syberfeldt, A.: Parameter tuning of MOEAs using a bilevel optimization approach. In: Gaspar-Cunha, A., Henggeler Antunes, C., Coello, C.C. (eds.) EMO 2015. LNCS, vol. 9018, pp. 233–247. Springer, Cham (2015). doi:10.1007/978-3-319-15934-8_16

    Google Scholar 

  2. Bader, J., Zitzler, E.: HypE: an algorithm for fast hypervolume-based many-objective optimization. Evol. Comput. 19(1), 45–76 (2011)

    Article  Google Scholar 

  3. Brockhoff, D., Tran, T., Hansen, N.: Benchmarking numerical multiobjective optimizers revisited. In: GECCO, pp. 639–646 (2015)

    Google Scholar 

  4. Das, I., Dennis, J.E.: Normal-boundary intersection: a new method for generating the pareto surface in nonlinear multicriteria optimization problems. SIAM J. Optim 8(3), 631–657 (1998)

    Article  MathSciNet  MATH  Google Scholar 

  5. Deb, K., Agrawal, S., Pratap, A., Meyarivan, T.: A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE TEVC 6(2), 182–197 (2002)

    Google Scholar 

  6. Deb, K., Jain, H.: An evolutionary many-objective optimization algorithm using reference-point-based nondominated sorting approach, part I: solving problems with box constraints. IEEE TEVC 18(4), 577–601 (2014)

    Google Scholar 

  7. Durillo, J.J., Nebro, A.J., Luna, F., Alba, E.: On the effect of the steady-state selection scheme in multi-objective genetic algorithms. In: Ehrgott, M., Fonseca, C.M., Gandibleux, X., Hao, J.-K., Sevaux, M. (eds.) EMO 2009. LNCS, vol. 5467, pp. 183–197. Springer, Heidelberg (2009). doi:10.1007/978-3-642-01020-0_18

    Chapter  Google Scholar 

  8. Eiben, A.E., Hinterding, R., Michalewicz, Z.: Parameter control in evolutionary algorithms. IEEE TEVC 3(2), 124–141 (1999)

    Google Scholar 

  9. Huband, S., Hingston, P., Barone, L., While, R.L.: A review of multiobjective test problems and a scalable test problem toolkit. IEEE TEVC 10(5), 477–506 (2006)

    MATH  Google Scholar 

  10. Ishibuchi, H., Masuda, H., Nojima, Y.: Comparing solution sets of different size in evolutionary many-objective optimization. In: IEEE CEC, pp. 2859–2866 (2015)

    Google Scholar 

  11. Ishibuchi, H., Sakane, Y., Tsukamoto, N., Nojima, Y.: Evolutionary many-objective optimization by NSGA-II and MOEA/D with large populations. In: IEEE SMC, pp. 1758–1763 (2009)

    Google Scholar 

  12. Ishibuchi, H., Setoguchi, Y., Masuda, H., Nojima, Y.: How to compare many-objective algorithms under different settings of population and archive sizes. In: IEEE CEC, pp. 1149–1156 (2016)

    Google Scholar 

  13. Ishibuchi, H., Tsukamoto, N., Nojima, Y.: Evolutionary many-objective optimization: a short review. In: IEEE CEC, pp. 2419–2426 (2008)

    Google Scholar 

  14. Li, B., Li, J., Tang, K., Yao, X.: Many-objective evolutionary algorithms: a survey. ACM Comput. Surv. 48(1), 13 (2015)

    Article  Google Scholar 

  15. López-Ibáñez, M., Knowles, J., Laumanns, M.: On sequential online archiving of objective vectors. In: Takahashi, R.H.C., Deb, K., Wanner, E.F., Greco, S. (eds.) EMO 2011. LNCS, vol. 6576, pp. 46–60. Springer, Heidelberg (2011). doi:10.1007/978-3-642-19893-9_4

    Chapter  Google Scholar 

  16. Martínez, S., Aguirre, H.E., Tanaka, K., Coello, C.A.C.: On the low-discrepancy sequences and their use in MOEA/D for high-dimensional objective spaces. In: IEEE CEC, pp. 2835–2842 (2015)

    Google Scholar 

  17. Radulescu, A., López-Ibáñez, M., Stützle, T.: Automatically improving the anytime behaviour of multiobjective evolutionary algorithms. In: Purshouse, R.C., Fleming, P.J., Fonseca, C.M., Greco, S., Shaw, J. (eds.) EMO 2013. LNCS, vol. 7811, pp. 825–840. Springer, Heidelberg (2013). doi:10.1007/978-3-642-37140-0_61

    Chapter  Google Scholar 

  18. Seada, H., Deb, K.: A unified evolutionary optimization procedure for single, multiple, and many objectives. IEEE TEVC 20(3), 358–369 (2016)

    Google Scholar 

  19. Smit, S.K., Eiben, A.E.: Parameter tuning of evolutionary algorithms: generalist vs. specialist. In: Chio, C., Cagnoni, S., Cotta, C., Ebner, M., Ekárt, A., Esparcia-Alcazar, A.I., Goh, C.-K., Merelo, J.J., Neri, F., Preuß, M., Togelius, J., Yannakakis, G.N. (eds.) EvoApplications 2010. LNCS, vol. 6024, pp. 542–551. Springer, Heidelberg (2010). doi:10.1007/978-3-642-12239-2_56

    Chapter  Google Scholar 

  20. Tan, Y., Jiao, Y., Li, H., Wang, X.: MOEA/D + uniform design: a new version of MOEA/D for optimization problems with many objectives. Comput. OR 40(6), 1648–1660 (2013)

    Article  MathSciNet  MATH  Google Scholar 

  21. Tušar, T., Filipič, B.: Differential evolution versus genetic algorithms in multiobjective optimization. In: Obayashi, S., Deb, K., Poloni, C., Hiroyasu, T., Murata, T. (eds.) EMO 2007. LNCS, vol. 4403, pp. 257–271. Springer, Heidelberg (2007). doi:10.1007/978-3-540-70928-2_22

    Chapter  Google Scholar 

  22. Wagner, T., Beume, N., Naujoks, B.: Pareto-, aggregation-, and indicator-based methods in many-objective optimization. In: Obayashi, S., Deb, K., Poloni, C., Hiroyasu, T., Murata, T. (eds.) EMO 2007. LNCS, vol. 4403, pp. 742–756. Springer, Heidelberg (2007). doi:10.1007/978-3-540-70928-2_56

    Chapter  Google Scholar 

  23. Wessing, S., Beume, N., Rudolph, G., Naujoks, B.: Parameter tuning boosts performance of variation operators in multiobjective optimization. In: Schaefer, R., Cotta, C., Kołodziej, J., Rudolph, G. (eds.) PPSN 2010. LNCS, vol. 6238, pp. 728–737. Springer, Heidelberg (2010). doi:10.1007/978-3-642-15844-5_73

    Google Scholar 

  24. Xiang, Y., Zhou, Y., Li, M., Chen, Z.: A vector angle based evolutionary algorithm for unconstrained many-objective optimization. IEEE TEVC (2016, in press)

    Google Scholar 

  25. Yuan, Y., Xu, H., Wang, B.: An experimental investigation of variation operators in reference-point based many-objective optimization. In: GECCO, pp. 775–782 (2015)

    Google Scholar 

  26. Yuan, Y., Xu, H., Wang, B., Yao, X.: A new dominance relation-based evolutionary algorithm for many-objective optimization. IEEE TEVC 20(1), 16–37 (2016)

    Google Scholar 

  27. Zhang, Q., Liu, W., Li, H.: The performance of a new version of MOEA/D on CEC09 unconstrained MOP test instances. In: IEEE CEC, pp. 203–208 (2009)

    Google Scholar 

  28. Zitzler, E., Thiele, L., Laumanns, M., Fonseca, C.M., da Fonseca, V.G.: Performance assessment of multiobjective optimizers: an analysis and review. IEEE TEVC 7(2), 117–132 (2003)

    Google Scholar 

Download references

Acknowledgments

This research is supported by the HPCI System Research Project “Research and development of multiobjective design exploration and high-performance computing technologies for design innovation” (Project ID:hp160203).

Author information

Authors and Affiliations

  1. Institute of Space and Astronautical Science, Japan Aerospace Exploration Agency, Sagamihara, Kanagawa, Japan

    Ryoji Tanabe & Akira Oyama

Authors
  1. Ryoji Tanabe
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Akira Oyama
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Ryoji Tanabe .

Editor information

Editors and Affiliations

  1. University of Münster, Münster, Germany

    Heike Trautmann

  2. TU Dortmund University, Dortmund, Germany

    Günter Rudolph

  3. University of Wuppertal, Wuppertal, Germany

    Kathrin Klamroth

  4. CINVESTAV-IPN, Mexico City, Mexico

    Oliver Schütze

  5. Clemson University, Clemson, South Carolina, USA

    Margaret Wiecek

  6. University of Surrey, Guildford, United Kingdom

    Yaochu Jin

  7. University of Münster, Münster, Germany

    Christian Grimme

Rights and permissions

Reprints and permissions

Copyright information

© 2017 Springer International Publishing AG

About this paper

Cite this paper

Tanabe, R., Oyama, A. (2017). The Impact of Population Size, Number of Children, and Number of Reference Points on the Performance of NSGA-III. In: Trautmann, H., et al. Evolutionary Multi-Criterion Optimization. EMO 2017. Lecture Notes in Computer Science(), vol 10173. Springer, Cham. https://doi.org/10.1007/978-3-319-54157-0_41

Download citation

  • DOI: https://doi.org/10.1007/978-3-319-54157-0_41

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-319-54156-3

  • Online ISBN: 978-3-319-54157-0

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics

Search

Navigation