Abstract
Creating ways for neural networks to evolve is an important goal in the field of artificial evolution. The main problem is how to encode the structure and properties of the neural network in the genome. If one overloads the genome with detailed network information, the evolutionary time increases prohibitively. If the genome is too simple, only simple problems can be solved. Since nature has found an efficient evolutionary solution to this problem, it is worth imitating the mechanisms by which biological neural nets are generated. In this article, a model is proposed in which artificial genomes increase the ability of axons to find, deteet, and connect to specific targets. Some initial simulation results for simple tasks are evolved, and the genetic tuning of the developmental processes of artificial evolution is discussed.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Edelman GM (1987) Neural Darwinism. Basic Books, New York
Harris WA, Holt CA (1999) Slit, the midline repellent. Naturé 398:462–463.
Gruau F, Whitley D (1993) The cellular development of neural networks: the interaction of learning and evolution. Technical Report 93-04, Laboratoire de l’Informatique du Parallélisme, Ecole Normale Supérieure de Lyon
Fleischer K, Barr AH (1994) A simulation test-bed for the study of multicellular development: the multiple mechanisms of morphogenesis. In: Langton CG (ed) Artificial life. III, Proceedings of the Workshop on Artificial Life. Addison-Wesley, Reading, p 389–416.
Vaario J, Shimohara K (1997) Synthesis of developmental and evolutionary modeling of adaptive autonomousl agents. In: Gerstner W, Germond A, Haster M, Nicoud J-D (eds) 7th International Conference of ARtificial Neural Networks (ICANN’97), Lausanne, 1997. Springer, p 721–726
Nolfi S, Floreano D, Miglino O, et al. (1994) How to evolve automonous robots: different approaches in evolutionary robotics. In: Meas P, Brooks RA (eds) Artificial life. IV. MIT Press, Cambridge
Dellaert F, Beer RD (1996) A developmental model for the evolution of complete autonomous agents. In: Macs P, Mataric MJ, Meyer J-A, Pollack J, Wilson SW (eds) From animals to animats. 4. Proceedings of the 4th International Conference of Simulation of Adaptive Behävior. MIT Press, Cambridge, p. 393–401
Eggenberger P (1997) Creation of neural networks based on developmental and evolutionary principles. In: Gerstner W, Germond A, Hasler M, Nicoud J-D (eds) 7th International Conference of Artificial Neural Networks (ICANN’97) Lausanne, 1997. Springer, p. 337–342
Eggenberger P, Dravid R (1999) An evolutionary approach to pattern formation mechanisms on lepidopteran wings. Congress of Evolutionary Computation, Washington, p 337–342
Gerhart J, Kirschner M (1997) Cells, embryos and evolution: toward a cellular and developmental understanding of phenotypic variation and evolutionary adaptability. Blackwell, Oxford
Author information
Authors and Affiliations
Corresponding author
Additional information
This work was presented in part at the Fifth International Symposium on Artificial Life and Robotics, Oita, Japan, January 26–28, 2000
About this article
Cite this article
Eggenberger, P. Axonal growth in evolutionary neurogenesis. Artif Life Robotics 5, 137–141 (2001). https://doi.org/10.1007/BF02481459
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF02481459