Abstract
Specialist and generalist behaviors in populations of artificial neural networks are studied. A genetic algorithm is used to simulate evolution processes, and thereby to develop neural network control systems that exhibit specialist or generalist behaviors according to the fitness formula. With evolvable fitness formulae the evaluation measure is let free to evolve, and we obtain a co-evolution of the expressed behavior and the individual evolvable fitness formula. The use of evolvable fitness formulae lets us work in a dynamic fitness landscape, opposed to most work, that traditionally applies to static fitness landscapes, only. The role of competition in specialization is studied by letting the individuals live under social conditions in the same, shared environment and directly compete with each other. We find, that competition can act to provide population diversification in populations of organisms with individual evolvable fitness formulae.
Keywords
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.
Preview
Unable to display preview. Download preview PDF.
References
Bernays, E. A. and Wcislo, W. T. (1994). Sensory Capabilities, Information Processing, and Resource Specialization. The Quarterly Review of Biology, 69, 2, pp. 187–204.
Brooker, L. C. and Brooker, M. G. (1990). Why are cuckoos host specific? Oikos, 57, pp. 301–309.
Futuyma, D. J. (1991). Evolution of host specificity in herbivorous insects: genetic, ecological, and phylogenetic aspects. In P. W. Price, T. M. Lewinsohn, G. W. Fernandes, and W. W. Benson (eds.) Plant-Animal Interactions, pp. 431–454, John Wiley & Sons, New York.
Goldberg, D. E. (1989). Genetic Algorithms in Search, Optimization and Machine Learning, Addison-Wesley, Reading.
Harris, M. and Ross, E.B. (1987). Food and Evolution. Toward a theory of human food habits. Philadelphia, Temple University Press.
Holland, J. J. (1975). Adaptation in natural and artificial systems. University of Michigan Press, Ann Arbor, Michigan.
Lund, H. H. (1994). Adaptive Approaches Towards Better GA Performance in Dynamic Fitness Landscapes. DAIMI PB-487, Aarhus University, Aarhus.
Lund, H. H. and Mayoh, B. H. (1995). Specialization in Populations of Artificial Neural Networks. In Proceedings of 5.th Scandinavian Conference on Artificial Intelligence (SCAI'95), IOS Press, Amsterdam.
Lund, H. H. and Parisi, D. (1994a). Simulations with an Evolvable Fitness Formula. Technical Report PCIA-1-94, C.N.R., Rome.
Lund, H. H. and Parisi, D. (1994b). Pre-adaptation in Populations of Neural Networks Evolving in a Changing Environment. C.N.R., Rome. Submitted to Artificial Life.
Matsuda, R. (1987). Animal Evolution in Changing Environments. John Wiley & Sons, New York.
Nolfi, S., and Parisi, D. (1993). Self-selection of input stimuli for improving performance. In C. A. Bekey (ed.) Neural Networks and Robotics. Kluwer Academic Press.
Parisi, D., Cecconi, F., and Nolfi, S. (1990). ECONETS: neural networks that learn in an environment. Network, 1, pp. 149–168.
Rothschild, M., and Clay, T. (1961). Fleas, Flukes, and Cuckoos. Arrow Books Ltd., London.
Rothstein, S. I. (1990). A model system for coevolution: avian brood parasitism. Annual Rev. Ecol. Syst., 21, pp. 481–508.
Santibanez-H., G., and Lindemann, M. (1986). Introduction to the Physiopathology of Neurotic States. VEB George, Leipzig.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1995 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Lund, H.H. (1995). Specialization under social conditions in shared environments. In: Morán, F., Moreno, A., Merelo, J.J., Chacón, P. (eds) Advances in Artificial Life. ECAL 1995. Lecture Notes in Computer Science, vol 929. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-59496-5_319
Download citation
DOI: https://doi.org/10.1007/3-540-59496-5_319
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-59496-3
Online ISBN: 978-3-540-49286-3
eBook Packages: Springer Book Archive