Abstract
A hypothetical neural scheme is proposed that ensures efficient decision making by an animal searching for food in a maze. Only the general structure of the network is fixed; its quantitative characteristics are found by numerical optimization that simulates the process of natural selection. Selection is aimed at maximization of the expected number of descendants, which is directly related to the energy stored during the reproductive cycle. The main parameters to be optimized are the increments of the interneuronal links and the working-memory constants.
Similar content being viewed by others
References
M. Mangel and K. Clark, Dynamic Models in the Ecology of Behavior [Russian translation], Mir, Moscow (1992).
M. Mangel, J. Mat. Biol.,28, 237 (1990).
D. S. Olton, J. T. Becker, and G. E. Hendelman, Behav. Brain Sci.,2, 313 (1979).
J. Buresh, O. Bureshova, and D. P. Houston, Procedures and Basic Experiments for Study of the Brain and Behavior [in Russian], Vysshaya Shkola, Moscow (1991).
F. Wasserman, Neurocomputing [Russian translation], Mir, Moscow (1992).
J. J. Hopfield and D. W. Tank, Biolog. Cybern.,52, 141 (1985).
S. Bornholdt and D. Graudenz, Neural Networks,5, 327 (1992).
Additional information
Biology Department, Moscow State University, Moscow. Translated from Izvestiya Vysshikh Uchebnykh Zavedenii, Radiofizika, Vol. 37, No. 9, pp. 1162–1172, September, 1994.
Rights and permissions
About this article
Cite this article
Budilova, E.V., Terekhin, A.T. & Chepurnov, S.A. A genetic algorithm for optimization of neural network capable of learning to search for food in a maze. Radiophys Quantum Electron 37, 749–755 (1994). https://doi.org/10.1007/BF01039615
Received:
Issue Date:
DOI: https://doi.org/10.1007/BF01039615