Abstract
Several recent works have examined the effectiveness of using knowledge models to guide search algorithms in high dimensional spaces. It seems that it may be a promising way to tackle some difficult problem. The aim of such methods is to reach good solutions using simultaneously evolutionary search and knowledge guidance. The idea proposed in this paper is to use a bayesian network in order to store and apply the knowledge model and, as a consequence, to accelerate the search process. A traditional evolutionary algorithm is modified in order to allow the reuse of the capitalized knowledge. The approach has been applied to a problem of selection of project scenarios in a multi-objective context. A preliminary version of this method was presented at EA’ 07 conference [1]. An experimentation platform has been developed to validate the approach and to study different modes of knowledge injection. The obtained experimental results are presented.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Pitiot, P., Coudert, T., Geneste, L., Baron, C.: Improvement of Intelligent Optimization by an experience feedback approach. In: Monmarché, N., Talbi, E.-G., Collet, P., Schoenauer, M., Lutton, E. (eds.) EA 2007. LNCS, vol. 4926, pp. 316–327. Springer, Heidelberg (2008)
Li, B., Chen, L., Huang, Z., Zhong, Y.: Product configuration optimization using a multiobjective GA. I.J. of Adv. Manufacturing Technology 30, 20–29 (2006)
Baron, C., Rochet, S., Esteve, D.: GESOS: a multi-objective genetic tool for project management considering technical and non-technical constraints. In: IFIP World Computer Congress on Art. Intel. Applications and Innovations, AIAI (2004)
Michalski, R.S., Wojtusiak, J., Kaufman, K.A.: Intelligent Optimization via Learnable Evolution Model. In: 18th Conf. on Tools with Artificial Intelligence, pp. 332–335 (2006)
Huyet, A.-L., Paris, J.-L.: Synergy between Evolutionary Optimization and Induction Graphs Learning for Simulated Manufacturing Systems. Inter. J. of Production Research 42(20), 4295–4313 (2004)
Pelikan, M., Sastry, K., Goldberg, D.E.: Sporadic model building for efficiency enhancement of the hBOA. Genetic Programming and Evolvable Machines 9, 53–84 (2008)
Baluja, S.: Using a priori knowledge to create probabilistic models for optimization. Inter. J. of approximate reasoning 31(3), 193–220 (2002)
Schwarz, J., Ocenasek, J.: A problem knowledge-based evolutionary algorithm KBOA for hypergraph bisectioning. In: 4th Joint Conf. on Knowledge-Based Software Engineering, pp. 51–58. IOS Press, Amsterdam (2000)
Hauschild, M.W., Pelikan, M., Sastry, K., Goldberg, D.E.: Using previous models to bias structural learning in the hierarchical BOA. In: Proceedings of the 10th annual conference on Genetic and evolutionary computation, pp. 415–422 (2008)
Sebag, M., Schoenauer, M.: A rule based similarity measure. In: Wess, S., Richter, M., Althoff, K.-D. (eds.) EWCBR 1993. LNCS, vol. 837, pp. 119–130. Springer, Heidelberg (1994)
Chung, C.J.: Knowledge based approaches to self adaptation in cultural algorithms. PhD thesis, Wayne State University, Detroit, USA (1997)
Zitzler, E., Thiele, L.: Multi objective EA: a comparative case study and the strength Pareto approach. IEEE Trans. on evolutionary computation 3(4), 257–271 (1999)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2010 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Pitiot, P., Coudert, T., Geneste, L., Baron, C. (2010). A Priori Knowledge Integration in Evolutionary Optimization. In: Collet, P., Monmarché, N., Legrand, P., Schoenauer, M., Lutton, E. (eds) Artifical Evolution. EA 2009. Lecture Notes in Computer Science, vol 5975. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-14156-0_9
Download citation
DOI: https://doi.org/10.1007/978-3-642-14156-0_9
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-14155-3
Online ISBN: 978-3-642-14156-0
eBook Packages: Computer ScienceComputer Science (R0)