Revising the Trade-off between the Number of Agents and Agent Intelligence

  • Marcus Komann
  • Dietmar Fey
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6024)


Emergent agents are a promising approach to handle complex systems. Agent intelligence is thereby either defined by the number of states and the state transition function or the length of their steering programs. Evolution has shown to be successful in creating desired behaviors for such agents. Genetic algorithms have been used to find agents with fixed numbers of states and genetic programming is able to balance between the steering program length and the costs for longer programs. This paper extends previous work by further discussing the relationship between either using more agents with less intelligence or using fewer agents with higher intelligence. Therefore, the Creatures’ Exploration Problem with a complex input set is solved by evolving emergent agents. It shows that neither a sole increase in intelligence nor amount is the best solution. Instead, a cautious balance creates best results.


State Machine Genetic Programming Cellular Automaton Agent Intelligence Visit Rate 
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.
    Anderson, C.: Creation of desirable complexity: strategies for designing selforganized systems. In: Complex Engineered Systems, pp. 101–121. Perseus Books Group (2006)Google Scholar
  2. 2.
    Beckers, R., Deneubourg, J.L., Goss, S.: Trails and U-turns in the selection of the shortest path by the ant Lasius Niger. Journal of Theoretical Biology 159, 397–415 (1992)CrossRefGoogle Scholar
  3. 3.
    Branke, J., Schmeck, H.: Evolutionary design of emergent behavior. In: Organic Computing, March 2008, pp. 123–140. Springer, Heidelberg (2008), CrossRefGoogle Scholar
  4. 4.
    Deb, K., Pratap, A., Agarwal, S., Meyarivan, T.: A fast and elitist multiobjective genetic algorithm: Nsga-ii. IEEE Transactions on Evolutionary Computation 6(2), 182–197 (2002)CrossRefGoogle Scholar
  5. 5.
    Deb, K.: A robust evolutionary framework for multi-objective optimization. In: GECCO 2008: Proceedings of the 10th annual conference on Genetic and evolutionary computation, pp. 633–640. ACM, New York (2008)CrossRefGoogle Scholar
  6. 6.
    Edmonds, B.: Using the experimental method to produce reliable self-organised systems. In: Brueckner, S.A., Di Marzo Serugendo, G., Karageorgos, A., Nagpal, R. (eds.) ESOA 2005. LNCS (LNAI), vol. 3464, pp. 84–99. Springer, Heidelberg (2005)Google Scholar
  7. 7.
    Goss, S., Aron, S., Deneubourg, J., Pasteels, J.: Self-organized shortcuts in the argentine ant. Naturwissenschaften 76(12), 579–581 (1989), CrossRefGoogle Scholar
  8. 8.
    Halbach, M.: Algorithmen und Hardwarearchitekturen zur optimierten Aufzählung von Automaten und deren Einsatz bei der Simulation künstlicher Kreaturen. Ph.D. thesis, Technische Universität Darmstadt (2008)Google Scholar
  9. 9.
    Halbach, M., Heenes, W., Hoffmann, R., Tisje, J.: Optimizing the behavior of a moving creature in software and in hardware. In: Sloot, P.M.A., Chopard, B., Hoekstra, A.G. (eds.) ACRI 2004. LNCS, vol. 3305, pp. 841–850. Springer, Heidelberg (2004)Google Scholar
  10. 10.
    Halbach, M., Hoffmann, R., Both, L.: Optimal 6-state algorithms for the behavior of several moving creatures. In: El Yacoubi, S., Chopard, B., Bandini, S. (eds.) ACRI 2006. LNCS, vol. 4173, pp. 571–581. Springer, Heidelberg (2006)CrossRefGoogle Scholar
  11. 11.
    Komann, M., Ediger, P., Fey, D., Hoffmann, R.: On the effectiveness of evolution compared to time-consuming full search of optimal 6-state automata. In: Vanneschi, L., Gustafson, S., Moraglio, A., De Falco, I., Ebner, M. (eds.) EuroGP 2009. LNCS, vol. 5481, pp. 280–291. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  12. 12.
    Komann, M., Fey, D.: Realising emergent image preprocessing tasks in cellular-automaton-alike massively parallel hardware. International Journal of Parallel, Emergent and Distributed Systems 22(2), 79–89 (2007)zbMATHCrossRefMathSciNetGoogle Scholar
  13. 13.
    Komann, M., Fey, D.: Evaluating the evolvability of emergent agents with different numbers of states. In: GECCO, pp. 1777–1778. ACM, New York (2009)CrossRefGoogle Scholar
  14. 14.
    Koza, J.R.: Scalable learning in genetic programming using automatic function definition, pp. 99–117 (1994)Google Scholar
  15. 15.
    Michalewicz, Z.: Genetic algorithms + data structures = evolution programs, 3rd edn. Springer, London (1996)zbMATHGoogle Scholar
  16. 16.
    Müller-Schloer, C., Sick, B.: Emergence in Organic Computing systems: Discussion of a controversial concept. In: Yang, L.T., Jin, H., Ma, J., Ungerer, T. (eds.) ATC 2006. LNCS, vol. 4158, pp. 1–16. Springer, Heidelberg (2006)CrossRefGoogle Scholar
  17. 17.
    Roska, T., Chua, L.: The cnn universal machine: an analogic array computer. IEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing 40(3), 163–173 (1993)zbMATHCrossRefMathSciNetGoogle Scholar
  18. 18.
    Wolf, T.D., Holvoet, T.: Emergence versus self-organisation: Different concepts but promising when combined. In: Brueckner, S.A., Di Marzo Serugendo, G., Karageorgos, A., Nagpal, R. (eds.) ESOA 2005. LNCS (LNAI), vol. 3464, pp. 1–15. Springer, Heidelberg (2005)Google Scholar
  19. 19.
    Wolfram, S.: A New Kind of Science. Wolfram Media Inc., Champaign (2002)zbMATHGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Marcus Komann
    • 1
  • Dietmar Fey
    • 2
  1. 1.Friedrich-Schiller-University JenaJenaGermany
  2. 2.University of Erlangen-NürnbergErlangenGermany

Personalised recommendations