Teaching Relaxation Labeling Processes Using Genetic Algorithms

  • Marcello Pelillo
  • Fabio Abbattista
  • Angelo Maffione
Conference paper


In a recent work, a learning procedure for relaxation labeling algorithms has been introduced which involves minimizing a certain cost function with classical gradient methods. The gradient-based learning algorithm suffers from some inherent drawbacks that could prevent its application to real-world problems of practical interest. Essentially, these include the inability to escape from local minima and its computational complexity. In this paper, we propose using genetic algorithms to solve the relaxation labeling learning problem to overcome the difficulties with the gradient algorithm. Experiments are presented which demonstrate the superiority of the proposed approach both in terms of quality of solutions and robustness.


Genetic Algorithm Gradient Algorithm Label Assignment Genetic Algorithm Genetic Algorithm Compatibility Model 
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.
    A. Rosenfeld, et al: IEEE-SMC 6, 420 (1976).MathSciNetMATHGoogle Scholar
  2. 2.
    R. Hummel, S. Zucker: IEEE-PAMI 5, 267 (1983).MATHCrossRefGoogle Scholar
  3. 3.
    L. Davis, A. Rosenfeld: Artif. Intell. 17, 245 (1981).CrossRefGoogle Scholar
  4. 4.
    S. Peleg, A. Rosenfeld: IEEE-SMC 8, 548 (1978).Google Scholar
  5. 5.
    M. Pelillo, M. Refice: IEEE-PAMI 16, 933 (1994).CrossRefGoogle Scholar
  6. 6.
    S. Zucker, et al: IEEE-PAMI 3, 117 (1981).MATHCrossRefGoogle Scholar
  7. 7.
    D. Goldberg: Genetic algorithms in search, optimization and machine learning. Addison-Wesley (1989).MATHGoogle Scholar
  8. 8.
    J. Lin: IEEE-IT 37, 145 (1991).MATHCrossRefGoogle Scholar
  9. 9.
    J. Grefenstette: IEEE-SMC 16, 122 (1986).Google Scholar
  10. 10.
    T. Caudell, C. Dolan: Proc. 3rd ICGA, 370 (1989).Google Scholar

Copyright information

© Springer-Verlag/Wien 1995

Authors and Affiliations

  • Marcello Pelillo
    • 1
  • Fabio Abbattista
    • 1
  • Angelo Maffione
    • 1
  1. 1.Dipartimento di InformaticaUniversità di BariBariItaly

Personalised recommendations