g-lvq, a combination of genetic algorithms and lvq

  • J. J. Merelo
  • A. Prieto


One of the ultimate goal in neural network research is the complete optimization of a neural network: topology, learning algorithm, initial weights and number of neurons. Up to now, only partial solutions have been found. This optimization should look at two conditions if the task assigned to the NN is going to be classification: accuracy, and obtention of a good representation of the sample. lvq neural nets are a supervised classification algorithms created by Kohonen. In this work, lvq NNs will be optimized using a GA according to three criteria: accuracy, net size and distortion. These three criteria are considered a vector fitness with three components that must be optimized separately. In order to carry this out, variable-length genomes are used to represent the neural network; each neuron is codified together with its label. Results in synthetic and real-world problems show that g-lvq is able to find an optimal size of the network, as well as combinations of weights that maximize classification accuracy.


Genetic Algorithm Weight Vector Maximize Classification Accuracy Grow Cell Structure Neural Network Research 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    Cliff, D.; Harvey, I.; Husbands, P.; Explorations in Evolutionary Robotics, Adaptive Behavior, vol. 2, no. 1 1993, 32 pps.Google Scholar
  2. [2]
    Nolfi, S.; Parisi, D.; Growing Neural Networks, Tech. Rep. PCIA-91-15, Department of Cognitive Processes and Artificial Intelligence, Institute of Psychology, CNR, Rome.Google Scholar
  3. [3]
    Gruau, F.; Genetic Synthesis of Modular Neural Networks, Procs. of the 5th Int. Conf. on Genetic Algorithms, S. Forrest, Ed., pp. 318 ff; 1993.Google Scholar
  4. [4]
    Harp, S.; Samad, T., Guha, A., Towards the genetic synthesis of neural networks, D.J. Schaffer (ed.), 3rd Int. Conf on Genetic Algorithms, pp. 360–369, 1989.Google Scholar
  5. [5]
    Kitano, H.; Designing neural network using genetic algorithm with graph generation system, Complex Systems, 4: 461–476, 1992.Google Scholar
  6. [6]
    Polani, D.; Uthmann, T., Training Kohonen Feature Maps in different Topologies: an Analysis using Genetic Algorithms, Procs. of the 5th Int. Conf. on Genetic Al-gorithms, S. Forrest, Ed., pp. 326-333, 1993.Google Scholar
  7. [7]
    Fritzke, B.; Growing Cell Structures: a Self-organizing Network in k Dimensions, Artificial Neural Networks 2, Aleksander, Taylor, Eds, 1993, pp 1051–1056.Google Scholar
  8. [8]
    Kohonen, T.; The Self-Organizing Map, Procs. IEEE, 78, 1464 ss, 1990.CrossRefGoogle Scholar
  9. [9]
    Duda, R. O., Hart, P. E., Pattern classification and scene analysis, New York: Wiley & Sons, 1973.MATHGoogle Scholar
  10. [10]
    Monte, E.; Hidalgo, D.; Marino, J.; Hernâez, I., A vector quantization algorithm based on genetic algorithms and LVQ, NATO-ASI Bubion 93, pp 231 ss, 1993.Google Scholar
  11. [11]
    Yao, X.; A Review of Evolutionary Artificial Neural Networks, Technical Report, 1992Google Scholar
  12. [12]
    Bengio, Y.; Bengio, S.; Learning a synaptic learning rule, Technical Report 751, Department d Informatique et de Recherche Operationelle, Université de Montreal, Canada, November 1990.Google Scholar
  13. [13]
    J. J. Merelo, Study, applications and optimization of neural vector quantization algorithms using genetic algorithms, PhD Thesis, University of Granada, Spain (in Spanish).Google Scholar
  14. [14]
    Goldberg, D.; Zen and the art of Genetic Algorithms, Procs ICGA89, 3rd Int. Conf. on Genetic Algorithms, 1989, J. D. Schaffer, Ed. pp 80–85,.Google Scholar
  15. [15]
    Lutz Pretchelt, PROBENI, A Set of Neural Network Benchmark Problems and Benchmarking Rules, Technical Report 21/94, Univ. Karlsruhe, Germany.Google Scholar
  16. [16]
    William H. Wolberg and O.L. Mangasarian, Multisurface method of pattern separation for medical diagnosis applied to breast cytology Proceedings of the National Academy of Sciences, U.S.A., Volume 87, December 1990, pp 9193–9196.MATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag/Wien 1995

Authors and Affiliations

  • J. J. Merelo
    • 1
  • A. Prieto
    • 1
  1. 1.Department of Electronics and Computer Technology, Facultad de CienciasCampus FuentenuevaGranadaSpain

Personalised recommendations