Designing Development Rules for Artificial Evolution

  • A. G. Rust
  • R. Adams
  • S. George
  • H. Bolouri


Using artificial evolution to successfully create neural networks requires appropriate developmental algorithms. The aim is to determine the least complex set of rules that allow a range of networks to evolve. This paper presents a set of generic growth rules that abstractly model the biological processes associated with the development of neuron-to-neuron connections. Substantially different 3D artificial neural structures can be grown by changing parameter values associated with the rules. A genetic algorithm has been successfully employed in determining parameter values that lead to specific neural structures.


Genetic Algorithm Bipolar Cell Cone Cell Artificial Evolution Genetic Algorithm Result 
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]
    E.J.W. Boers, H. Kuiper, B.L.M. Happel, and I.G. Sprinkhuizen-Kuyper. Designing modular artificial neural networks. Technical Report 93-24, Dept. of Computer Science, Leiden University, Netherlands, September 1993.Google Scholar
  2. [2]
    J. Branke. Evolutionary algorithms for neural network design and training. In Proceedings of the 1st Nordic Workshop on Genetic Algorithms and its Applications, 1995.Google Scholar
  3. [3]
    F. Dellaert and R.D. Beer. A Developmental Model for the Evolution of Complete Autonomous Agents. MIT Press / Bradford Books, 1996.Google Scholar
  4. [4]
    K. Fleischer. A Multiple-Mechanism Developmental Model for Defining Self-Organizing Geometric Structures. PhD thesis, California Institute of Technology, May 1995.Google Scholar
  5. [5]
    J.J. Grefenstette. Genesis 5.0. Scholar
  6. [6]
    J. Hertz, A. Krogh, and R.P. Palmer. Introduction to the Theory of Neural Computation. Addison Wesley, Redwood, CA, 1991.Google Scholar
  7. [7]
    J. Kodjabachian and J-A. Meyer. Evolution and development of control architectures in animats. Robotics and Autonomous Systems. 16:161–182, 1995.CrossRefGoogle Scholar
  8. [8]
    M. Mataric and D. Cliff. Challenges in evolving controllers for physical robots. Technical Report CS-95-184, Brandeis University, USA, November 1995.Google Scholar
  9. [9]
    B.W. Mel. Information-processing in dendritic trees. Neural Computation, 6(6):1031–1085, 1994.MATHCrossRefGoogle Scholar
  10. [10]
    A.G. Rust, S. George, H. Bolouri, and R. Adams. Developmental artificial neural networks for shape recognition: A model of the retina. Technical Report Technical Memorandum ERDC/1996/0011, ERDC, University of Hertfordshire, UK, May 1996.Google Scholar
  11. [11]
    C.J. Shatz. Role for spontaneous neural activity in the patterning ofconnections between retina and LGN during visual system development. International Journal of Developmental Neuroscience, 12(6):531–546, 1994.CrossRefGoogle Scholar
  12. [12]
    M. P. Stryker. Precise development from imprecise rules. Science, 263:1244–1245, 4 March 1994.CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Wien 1998

Authors and Affiliations

  • A. G. Rust
    • 1
    • 2
  • R. Adams
    • 1
  • S. George
    • 1
  • H. Bolouri
    • 2
    • 3
  1. 1.Division of Computer ScienceUniversity of HertfordshireUK
  2. 2.Engineering Research and Development CentreUniversity of HertfordshireUK
  3. 3.Biology 216-76California Institute of TechnologyUSA

Personalised recommendations