Abstract
In this paper, we consider each neural network as a point in a multi-dimensional problem space and suggest a crossover that locates the central point of a number of neural networks. By this, genetic algorithms can spend more time around attractive areas. We also apply representational normalization to neural networks to maintain genotype consistency in crossover. For the normalization, we utilize the Hungarian method of matching problems. The experimental results of our neuro-genetic algorithm overall showed better performance over the traditional multi-start heuristic and the genetic algorithm with a traditional crossover. These results are evidence that it is attractive to exploit central areas of local optima.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Belew, R.K., McInerney, J., Schraudolph, N.N.: Evolving networks: Using the genetic algorithm with connectionist learning. In: Langton, C.G., Taylor, C., Farmer, J.D., Rasmussen, S. (eds.) Artificial Life II, pp. 511–547. Addison- Wesley, Redwood City (1992)
Blake, C.L., Merz, C.J.: UCI repository of machine learning databases (1998)
Boese, K.D., Kahng, A.B., Muddu, S.: A new adaptive multi-start technique for combinatorial global optimizations. Operations Research Letters 15, 101–113 (1994)
Bui, T.N., Moon, B.R.: A new genetic approach for the traveling salesman problem. In: IEEE Conference on Evolutionary Computation, pp. 7–12 (1994)
Chalmers, D.J.: The evolution of learning: An experiment in genetic connectionism. In: Proceedings of the 1990 Connectionist Models Summer School, pp. 81–90 (1990)
Choi, S.S., Moon, B.R.: Normalization in genetic algorithms. In: Genetic and Evolutionary Computation Conference, pp. 862–873 (2003)
Crosher, D.: The artificial evolution of a generalized class of adaptive processes. In: Preprints of AI 1993 Workshop on Evolutionary Computation, pp. 18–36 (1993)
Gale, D.: The Theory of Linear Economic Models. McGraw-Hill Book Company, Inc, New York (1960)
Hancock, P.J.B.: Genetic algorithms and permutation problems: a comparison of recombination operators for neural net structure specification. In: Proc. Int. Workshop Combinations of Genetic Algorithms and Neural Networks, pp. 108–122 (1992)
Haykin, S.: Neural Networks, A Comprehensive Foundation. Prentice-Hall, Englewood Cliffs (1999)
James, A., David, M.: Neural Networks, Algorithms, Applications, and Programming Techniques. Addison-Wesley, Reading (1994)
Jones, T., Forrest, S.: Fitness distance correlation as a measure of problem difficulty for genetic algorithms. To appear in Sixth International Conference on Genetic Algorithms (1995)
Kauffman, S.: Adaptation on rugged fitness landscapes. Lectures in the Science of Complexity, 527–618 (1989)
Kim, H.B., Jung, S.H., Kim, T.G., Park, K.H.: Fast learning method for backpropagation neural network by evolutionary adaptation of learning rates. Neurocomputating
Kim, Y.H., Moon, B.R.: Investigation of the fitness landscapes and multi-parent crossover for graph bipartitioning. In: Genetic and Evolutionary Computation Conference, pp. 1123–1134 (2003)
Koza, J.R., Rice, J.P.: Genetic generation of both the weights and architecture for a neural network. In: IEEE Int. Joint Conf. Neural Networks, pp. 71–76 (1991)
Kuhn, H.W.: The Hungarian method for the assignment problem. Naval Res. Naval Res. Logist. Quart. 2, 83–97 (1955)
Lin, C.T., Jou, C.P.: Controlling chaos by GA-based reinforcement learning neural network. IEEE Trans. on Neural Networks 10(4), 846–869 (1999)
Liu, Y., Yao, X.: Evolutionary design of artificial neural networks with different nodes. In: IEEE Conference on Evolutionary Computation, pp. 670–675 (1996)
Manderick, B.,, M.: deWeger, and P. Spiessens. The genetic algorithm and the structure of the fitness landscape. In: International Conference on Genetic Algorithms, pp. 143–150 (1991)
Mangasarian, O.L., Wolberg, W.H.: Cancer diagnosis via linear programming. SIAM News 23(5), 1–18 (1990)
Michalewicz, Z.: Genetic Algorithms + Data Structure = Evolution Programs. Springer, Berlin (1992)
Montana, D., Davis, L.: Training feedforward neural network using genetic algorithms. In: 11th International Joint Conference on Artificial Intelligence, pp. 762–767 (1989)
Petridis, V., Paterakis, E., Kehagias, A.: A hybrid neural-genetic multimodel parameter estimation algorithm. IEEE Trans. on Neural Networks 9(5), 862–876 (1998)
Pujol, J., Poli, R.: Evolving neural networks using a dual representation with a combined crossover operator. In: IEEE Conference on Evolutionary Computation, pp. 416–421 (1998)
Radcliffe, N.J.: Forma analysis and random respectful recombination. In: International Conference on Genetic Algorithms, pp. 222–229 (1991)
Sexton, R.S., Dorsey, R.E., Johnson, J.D.: Toward global optimization of neural networks: A comparison of the genetic algorithm and backpropagation. Decision Support Systems 22(2), 171–185 (1998)
Sorkin, G.B.: Efficient simulated annealing on fractal landscapes. Algorithmica 6, 367–418 (1991)
Thierens, D.: Non-redundant genetic coding of neural networks. In: IEEE Conference on Evolutionary Computation, pp. 571–575 (1996)
Weinberger, E.D.: Fourier and Taylor series on fitness landscapes. Biological Cybernetics 65, 321–330 (1991)
Whitley, D., Starkweather, T., Bogart, C.: Genetic algorithms and neural networks: Optimizing connections and connectivity. Parallel Computing 14(3), 347–361 (1990)
Yao, X.: Evolving artificial neural networks. Proceedings of the IEEE 87(9), 1423–1447 (1999)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Jung, S., Moon, BR. (2004). Central Point Crossover for Neuro-genetic Hybrids. In: Deb, K. (eds) Genetic and Evolutionary Computation – GECCO 2004. GECCO 2004. Lecture Notes in Computer Science, vol 3102. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-24854-5_124
Download citation
DOI: https://doi.org/10.1007/978-3-540-24854-5_124
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-22344-3
Online ISBN: 978-3-540-24854-5
eBook Packages: Springer Book Archive