Skip to main content
SpringerLink
Log in

Local Fitness Meta-Models with Nearest Neighbor Regression

  • Conference paper
  • First Online:
Applications of Evolutionary Computation (EvoApplications 2016)

Part of the book series: Lecture Notes in Computer Science ((LNTCS,volume 9598))

Included in the following conference series:

Abstract

In blackbox function optimization, the results of fitness function evaluations can be used to train a regression model. This meta-model can be used to replace function evaluations and thus reduce the number of fitness function evaluations in evolution strategies (ES). In this paper, we show that a reduction of the number of fitness function evaluations of a (1+1)-ES is possible with a combination of a nearest neighbor regression model, a local archive of fitness function evaluations, and a comparatively simple meta-model management. We analyze the reduction of fitness function evaluations on set of benchmark functions.

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

References

  1. Armstrong, M.: Basic Linear Geostatistics. Springer, Heidelberg (1998)

    Book  MATH  Google Scholar 

  2. Bouzarkouna, Z., Auger, A., Ding, D.Y.: Local-meta-model CMA-ES for partially separable functions. In: Genetic and Evolutionary Computation Conference (GECCO), pp. 869–876 (2011)

    Google Scholar 

  3. Cruz-Vega, I., Garcia-Limon, M., Escalante, H.J.: Adaptive-surrogate based on a neuro-fuzzy network and granular computing. In: Genetic and Evolutionary Computation Conference (GECCO), pp. 761–768 (2014)

    Google Scholar 

  4. Delgado, M.F., Cernadas, E., Barro, S., Amorim, D.G.: Do we need hundreds of classifiers to solve real world classification problems? J. Mach. Learn. Res. 15(1), 3133–3181 (2014)

    MathSciNet  MATH  Google Scholar 

  5. Elsayed, S.M., Ray, T., Sarker, R.A.: A surrogate-assisted differential evolution algorithm with dynamic parameters selection for solving expensive optimization problems. In: Proceedings of IEEE Congress on Evolutionary Computation (CEC), pp. 1062–1068 (2014)

    Google Scholar 

  6. Jin, Y., Olhofer, M., Sendhoff, B.: On evolutionary optimization with approximate fitness functions. In: Genetic and Evolutionary Computation Conference (GECCO), pp. 786–793 (2000)

    Google Scholar 

  7. Kramer, O., Schlachter, U., Spreckels, V.: An adaptive penalty function with meta-modeling for constrained problems. In: Proceedings of IEEE Congress on Evolutionary Computation (CEC), pp. 1350–1354 (2013)

    Google Scholar 

  8. Kruisselbrink, J.W., Emmerich, M.T.M., Deutz, A.H., Bäck, T.: A robust optimization approach using kriging metamodels for robustness approximation in the CMA-ES. In: Proceedings of IEEE Congress on Evolutionary Computation (CEC), pp. 1–8 (2010)

    Google Scholar 

  9. Liao, Q., Zhou, A., Zhang, G.: A locally weighted metamodel for pre-selection in evolutionary optimization. In: Proceedings of IEEE Congress on Evolutionary Computation (CEC), pp. 2483–2490 (2014)

    Google Scholar 

  10. Loshchilov, I., Schoenauer, M., Sebag, M.: A mono surrogate for multiobjective optimization. In: Genetic and Evolutionary Computation Conference (GECCO), pp. 471–478 (2010)

    Google Scholar 

  11. Loshchilov, I., Schoenauer, M., Sebag, M.: Self-adaptive surrogate-assisted covariance matrix adaptation evolution strategy. In: Genetic and Evolutionary Computation Conference (GECCO), pp. 321–328 (2012)

    Google Scholar 

  12. Loshchilov, I., Schoenauer, M., Sebag, M.: Intensive surrogate model exploitation in self-adaptive surrogate-assisted cma-es (saacm-es). In: Genetic and Evolutionary Computation Conference (GECCO), pp. 439–446 (2013)

    Google Scholar 

  13. Martínez, S.Z., Coello, C.A.C.: A multi-objective meta-model assisted memetic algorithm with non gradient-based local search. In: Genetic and Evolutionary Computation Conference (GECCO), pp. 537–538 (2010)

    Google Scholar 

  14. Preuss, M., Rudolph, G., Wessing, S.: Tuning optimization algorithms for real-world problems by means of surrogate modeling. In: Genetic and Evolutionary Computation Conference (GECCO), pp. 401–408 (2010)

    Google Scholar 

  15. Rechenberg, I.: Evolutionsstrategie - Optimierung technischer Systeme nach Prinzipien der biologischen Evolution. Frommann-Holzboog, Stuttgart (1973)

    Google Scholar 

  16. Rosales-Pérez, A., Coello, C.A.C., Gonzalez, J.A., García, C.A.R., Escalante, H.J.: A hybrid surrogate-based approach for evolutionary multi-objective optimization. In: Proceedings of IEEE Congress on Evolutionary Computation (CEC), pp. 2548–2555 (2013)

    Google Scholar 

  17. Verbeeck, D., Maes, F., Grave, K.D., Blockeel, H.: Multi-objective optimization with surrogate trees. In: Genetic and Evolutionary Computation Conference (GECCO), pp. 679–686 (2013)

    Google Scholar 

  18. Willmes, L., Bäck, T., Jin, Y., Sendhoff, B.: Comparing neural networks and kriging for fitness approximation in evolutionary optimization. In: Proceedings of IEEE Congress on Evolutionary Computation (CEC), pp. 663–670 (2003)

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Computational Intelligence Group, Department of Computing Science, University of Oldenburg, Oldenburg, Germany

    Oliver Kramer

Authors
  1. Oliver Kramer
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Oliver Kramer .

Editor information

Editors and Affiliations

  1. Politecnico di Torino, Turin, Italy

    Giovanni Squillero

  2. Aalborg University, Copenhagen, Denmark

    Paolo Burelli

A Benchmark Functions

A Benchmark Functions

In this work, we employ the following benchmark problems:

  • Sphere \(f(\mathbf {x}) = \sum _{i=1}^d (x_i)^2\)

  • Rosenbrock \(f(\mathbf {x})= \sum _{i=1}^{d-1}\left( 100(x_i^2-x_{i+1})^2+(x_i-1)^2\right) \)

Rights and permissions

Reprints and permissions

Copyright information

© 2016 Springer International Publishing Switzerland

About this paper

Cite this paper

Kramer, O. (2016). Local Fitness Meta-Models with Nearest Neighbor Regression. In: Squillero, G., Burelli, P. (eds) Applications of Evolutionary Computation. EvoApplications 2016. Lecture Notes in Computer Science(), vol 9598. Springer, Cham. https://doi.org/10.1007/978-3-319-31153-1_1

Download citation

  • DOI: https://doi.org/10.1007/978-3-319-31153-1_1

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-319-31152-4

  • Online ISBN: 978-3-319-31153-1

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics

Search

Navigation