A Research Based on a Modified Genetic Algorithm for the Overfitting of Resonance Maching Network

  • Changji Wen
  • Cuijuan Zhou
  • Shengsheng Wang
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 124)


During the category learning process of SFAM ,due to the inputting category sample series containing noises or samples overlapping happened in any way easily, the phenomenon results to network structure redundancy and generalization ability problem.Aimed at it, this passage proposed a modified learning strategy based on an adaptive genetic algorithm. Firstly,to the variable structure genome sequence encoded composed of network conjunction weights and category nodes,we carry out a new operator—cutting operator,Instead of crossover operator in traditional GA, the operator depending on the rules of credibility judgement,cuts down the redundant category nodes which appeared in overfitting phenomenon. Secondly,to the mutation operator,we adjust the mutation rate by population fitness average. This passage takes the SFAM as a prototype, propose an modified learning strategy based on adaptive genetic algorithm which compared with SFAM and other improved ssFAM, ssGAM, ssEAM, Safe-μARTMAP network models,the experiment results show that generalization ability has improved, and the modified network is obvious promoted in solving the overfitting phenomenon problem and reducing the network redundancy.


Genetic Algorithm Adaptive Resonance Theory Category Node Confidence Factor Adaptive Genetic Algorithm 
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.
    Lippmann, R.P.: An introduction to computing with neural nets. IEEE ASSP Magazine 4(2), 4–22 (1987)CrossRefGoogle Scholar
  2. 2.
    Kramer, M.A., Leonard, J.A.: Diagnosis using back-propagation neural networks:analysis and criticism. Computer Chen. Engin 14(12), 1328–1338 (1990)Google Scholar
  3. 3.
    Ripley, B.D.: Neural networks and related methods for classification. J. R. Statist. Soc. B 56(3), 409–456 (1994)zbMATHMathSciNetGoogle Scholar
  4. 4.
    Paetz, J.: Reducing the number of neurons inradial basis function networks with dynamic decay adjustment. Neurocomputing 62, 79–91 (2004)CrossRefGoogle Scholar
  5. 5.
    Tan, S.C., Rao, M.V.C., Lim, C.P.L.: On the reduction of complexity in the architect- ture of Fuzzy ARTMAP with dynamic decay adjustment. Neurocomputing 69, 2456–2460 (2006)CrossRefGoogle Scholar
  6. 6.
    Ye, Z., Li, Z., Xie, M.: Some improvements onada-ptive genetic algorithms forreliability related applications. Reliability Engineering and System Saftety 95(2), 120–126 (2010)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Changji Wen
    • 1
  • Cuijuan Zhou
    • 1
  • Shengsheng Wang
    • 2
  1. 1.School of Informationg TechnologyJilin Agricultrue UniversivyChangchunChina
  2. 2.College of Computer Science and TechnologyJilin UniversityChangchunChina

Personalised recommendations