Abstract
In orientation to new developments in evolutionary biology we propose an extension of evolutionary algorithms in two dimensions, the regulatory algorithm (RGA). It consists of two levels of vectors, the regulatory vector and the structural vector. Each element of the regulatory vector is connected with one or several elements of the structural vector, but not vice versa. The connections can be interpreted as steering connections, the switching on or off of the structural elements and/or as switching orders for the structural elements. An RGA that operates with the usual genetic operators of mutation and crossover can be used for avoiding rules like penalty or default operators, it is in certain problems significantly faster than a standard genetic algorithm, and it is very suited when modeling and optimizing systems that consist themselves of different levels. Examples of RGA usage are shown, namely, the optimal dividing of socially deviant youths in a hostel, the optimal introduction of communication standards in information systems, and the allocation of employees to superiors by taking into regard the different personality types.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Rechenberg and Holland were of course not the first who developed algorithms orientated to the paradigm of biological evolution but their work certainly founded the field of evolutionary algorithms as an autonomous field of research.
The available literature on biological evolution told us just little about the question if evolution operated chiefly on the RV level, the SV level, or even on the CV. Several geneticists whom we asked told us that according to their knowledge unfortunately genetics had never thoroughly dealt with this problem.
In the last years a new approach to use evolutionary algorithms has been proposed, namely the usage of so-called Multilevel Evolutionary Algorithms (MLEA) or more specifically Multilevel Genetic Algorithms (MLGA) (e.g. Soper et al. 2000; Antonio 2002). Despite the striking semantical similarity this approach is factually an inverse way: In MLEA a problem is decomposed in different “levels” and on each level an EA is applied. The algorithms remain one-dimensional in contrast to the RGA where the algorithm itself consists of different levels.
Readers who are interested in the socio-psychological details and are able to read German texts are referred to Klüver et al. (2006).
Readers who are interested in the technical details of these experiments (and are able to read German) might obtain from the authors the master thesis of our MSc student Silas Graffy who performed the different experiments.
The shell we mentioned in Sect. 3, called NAOP—nature analogous optimization procedures—contains subshells for genetic algorithms, evolution strategies, simulated annealing, and different versions of the RGA, in particular one version for a RGA that operates like a GA and a RGA that operates like an ES (cf. Klüver and Klüver 2013).
This idea was independently developed by our two Masters students, Silas Graffy and Markus Mejer.
References
Antonio C (2002) A multi level genetic algorithm for optimization of geometrically nonlinear stiffened composite structures. Struct Multidiscip Optim 24(5):372–386
Behe J (2008) The edge of evolution: the search for the limits of Darwinism. Free Press, New York
Carroll SB (2006) Endless forms most beautiful. the new science of evo eevo and the making of the animal kingdom. Weidenfeld and Nicolson, London
Grolle J (2009) Grammatik des Lebens. In: Der SPIEGEL 2009/4, pp 98–109
Holland JH (1975) Adaptation in natural and artificial systems. MIT Press, Cambridge
Huxley JS (1942) Evolution. The modern synthesis. Allen and Unwin, London
Jacob F, Monod J (1961) Genetic regulatory mechanisms in the synthesis of proteins. Mol Biol 3:318
Klüver C, Klüver J (2013) Soft computing tools for the analysis of complex systems. In: Proceedings of the 1st international conference SIMEX’2013, pp 23–30
Klüver J, Schmidt J, Stoica C (2006) Computersimulationen und soziale Einzelfallstudien (computer simulations and social case studies). w3l Verlag, Herdecke
Michalewicz Z (1994) Genetic algorithms + data structures = evolution programs. Springer, Berlin
Rechenberg I (1972) Evolutionsstrategien. Friedrich Frommann Verlag, Stuttgart
Soper AJ, Walshaw C, Cross M (2000) A combined evolutionary search and multi level optimisation approach to graph partitioning. University of Greenwich, Mathematics Research Report 00/IM/ 58
Acknowledgments
We thank an anonymous reviewer and Jörn Schmidt from our research group for giving us several helpful comments to improve an earlier version of this article.
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by V. Loia.
Rights and permissions
About this article
Cite this article
Klüver, J., Klüver, C. The regulatory algorithm (RGA): a two-dimensional extension of evolutionary algorithms. Soft Comput 20, 2067–2075 (2016). https://doi.org/10.1007/s00500-015-1624-6
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00500-015-1624-6