Deb, K., Goldberg, D. E.: An Investigation of Niche and Species Formation in Genetic Function Optimization. In: Schaffer, J. D. (ed.): Proceedings of Third International Conference of Genetic Algorithms, Morgan Kaufmann, San Mateo, CA (1989) 42-50  Home/Search   Document Not in Database   Summary   Related Articles   Check

....to handle constraints in single objective optimization problems. The NPGA is a multiobjective optimization approach in which individuals are selected through a tournament based on Pareto dominance. However, unlike the NPGA, Coello and Mezura s approach does not require niches (or fitness sharing [10]) to maintain diversity in the population. The NPGA is a more efficient technique than traditional multiobjective optimization algorithms, since it does not compare every individual in the population with respect to each other (as in traditional Pareto ranking) but uses only a sample of the ....

Kalyanmoy Deb and David E. Goldberg. An Investigation of Niche and Species Formation in Genetic Function Optimization. In J. David Schaffer, editor, Proceedings of the Third International Conference on Genetic Algorithms, pages 42--50, San Mateo, California, June 1989. George Mason University, Morgan Kaufmann Publishers.

....than one particle are assigned a tness equal to the result of dividing any number x 1 (we used x = 10 in our experiments) by the number of particles that they contain. This aims to decrease the tness of those hypercubes that contain more particles and it can be seen as a form of tness sharing [5]. Then, we apply roulettewheel selection using these tness values to select the hypercube from which we will take the corresponding particle. Once the hypercube has been selected, we select randomly a particle within such hypercube. POP [i] is the current value of the particle i. b) Compute the ....

Kalyanmoy Deb and David E. Goldberg. An Investigation of Niche and Species Formation in Genetic Function Optimization. In J. David Schaer, editor, Proceedings of the Third International Conference on Genetic Algorithms, pages 42-50, San Mateo, California, June 1989. George Mason University, Morgan Kaufmann Publishers.

....1. A. Fitness Assignment For each generation, all non dominated solutions of the # or # solutions will constitute the 1st front. To these solutions a fitness value of 1 is assigned. In order to maintain diversity, a sharing scheme is then applied to the fitness values of these solutions [1]. Thus, the fitness value of each solution is divided by a quantity, called niche count, proportional to the number of solutions having a distance inferior to a parameter, the # share . All distances are measured in objective space. Thereafter, the solutions of the 1st front are ignored ....

K. Deb and D. Goldberg. "An investigation of niche and species formation in genetic function optimization". Proceedings of the Third International Conference on Genetic Algorithms, pp. 42-50, 1989.

....the Pareto set may not be uniformly sampled. Usually, the finite populations will converge to only one or some of these, due to stochastic errors in the selection process. Such phenomenon is known as genetic drift. Therefore, the additional use of fitness sharing [5] and mating restriction [1] can be used to prevent the drift and promote the sampling of the whole Pareto set by the population. a) Mating Restriction: Mating restriction has been implemented based on the distance between individuals in the objective domain. The distance between chromosome and is defined as (9) Based on ....

K. Deb and D. E. Goldberg, "An investigation of niche and species formation in genetic function optimization," in Proceedings of the 3rd International Conference on Genetic Algorithms, J. D. Schaffer, Ed. San Mateo, CA: Morgan Kaufmann, 1989, pp. 42--50.

....to handle constraints in single objective optimization problems. The NPGA is a multiobjective optimization approach in which individuals are selected through a tournament based on Pareto dominance. However, unlike the NPGA, Coello and Mezura s approach does not require niches (or tnes sharing [16]) to maintain diversity in the population. The NPGA is a more ecient technique than traditional multiobjective optimization algorithms, since it does not compare every individual in the population with respect to each other (as in traditional Pareto ranking) but uses only a sample of the ....

Kalyanmoy Deb and David E. Goldberg. An Investigation of Niche and Species Formation in Genetic Function Optimization. In J. David Schaffer, editor, Proceedings of the Third International Conference on Genetic Algorithms, pages 42-50, San Mateo, California, June 1989. George Mason University, Morgan Kaufmann Publishers.

.... there are constraints very dicult to satisfy) this technique will fail unless a feasible point is introduced in the initial population [104] Deb s results [44] are very encouraging, but his technique seems to have problems to maintain diversity in the population, and the use of niching methods [45] combined with higher than usual mutation rates is apparently necessary to avoid stagnation. Sharing is an expensive process (O(n ) and its use introduces an extra parameter ( share ) whose de nition is normally determined using an empirical procedure similar to the one used with the other ....

Kalyanmoy Deb and David E. Goldberg. An Investigation of Niche and Species Formation in Genetic Function Optimization. In J. David Schaer, editor, Proceedings of the Third International Conference on Genetic Algorithms, pages 42-50, San Mateo, California, June 1989. George Mason University, Morgan Kaufmann Publishers.

....1 No Figure 4: Flow chart of NSGA. the maximum distance allowed between any two individuals to become members of a niche. In any application of sharing, we can implement either genotypic sharing, since we always have a genotype (the encoding) or phenotypic sharing. However, Deb and Goldberg in [5] indicate that in general, phenotypic sharing is superior to genotypic sharing. Thus, we have used a phenotypic sharing which is calculated from the normalized Euclidean distance between the objective functions. The parameter share can be calculated as follows [5] share 0:5 q (5) ....

....However, Deb and Goldberg in [5] indicate that in general, phenotypic sharing is superior to genotypic sharing. Thus, we have used a phenotypic sharing which is calculated from the normalized Euclidean distance between the objective functions. The parameter share can be calculated as follows [5]: share 0:5 q (5) where q is the desired number of distinct Pareto optimal solutions and p is the number of decision variables. Although the calculation of share depends on this parameter q, it has been shown [35] that the use of the above equation with q 10 works in many test ....

K. Deb and D. E. Goldberg. An investigation of niche and species formation in genetic function. In Proc. of 3 rd International Conference on Genetic Algorithms, pages 4250, 1989.

....used the one max problem, which is probably the most frequently used test function in research on genetic algorithms because of its simplicity. This function measures the fitness of an individual as the number of bits set to one on the chromosome. The parameter share can be calculated as follows [6]: share 0:5 q (1) where q is the desired number of distinct Pareto optimal solutions and p is the number of decision variables. Although the calculation of share depends on this parameter q, it has been shown [8] that the use of the above equation with q 10 works in many test ....

K.Deb and D.E.Goldberg. An investigation of niche and species formation in genetic function. In Proc. of 3 rd Conference on Genetic Algorithms, pages 42--50, 1989.

....diversity . Diversity is the average square distance to other members of the population, using a specialised measure of edit distance between nodes. This multiobjective method promotes smaller and more diverse trees. McKay [4] applies the traditional tness sharing concept from Deb and Goldberg [16] to test its feasibility in genetic programming. Diversity is the number of tness cases found, and the sharing concept assigns a tness based on an individual s performance divided by the number of other individuals with the same performance. McKay also studies negative correlation and a root ....

K. Deb and D.E. Goldberg. An investigation of niche and species formation in genetic function optimization. In J. D. Schaer, editor, Proceedings of the Third International Conference on Genetic Algorithms, Washington DC, 1989.

....constraint. This e#ect, where features come into balance with respect to resources consumed is another emergent consequence of selfish agents under reproductive plans. Some EC models have explicitly exploited this emergent resource balancing e#ect (e.g. in multi objective optimisation problems [Deb and Goldberg, 1989, Mahfoud, 1997] Such e#ects have also been observed as emergent artefacts of EC system dynamics [Horn et al. 1994, Smith et al. 1993] In the most general situation, the perceived utility of a given feature in one subpopulation of agents will be a function of the presence of other features in ....

Deb, K. and Goldberg, D. E. (1989). An investigation of niche and species formation in genetic function optimization. Proceedings of the Third International Conference on Genetic Algorithms, pages 42--50.

....in Step C2, each requiring O(N log N) computations. Step C3 requires N computations. Thus, the complexity of the above distance metric computation is O(MN log N ) For large N , this complexity is smaller than O(MN ) which denotes the computational complexity required in other niching methods [6]. 3.2 A Hybrid MOEA Like in single objective EAs, it has been observed that MOEAs (including the NSGA II) are capable of reaching close to the Pareto optimal front quickly, but then getting slow in converging to the true Pareto optimal front. In fact, a recent study has shown that the ....

K. Deb and D. E. Goldberg. An investigation of niche and species formation in genetic function optimization. In Proceedings of the Third International Conference on Genetic Algorithms, pages 42-50, 1989.

No context found.

Deb, K. and Goldberg, D. E. (1989). An investigation of niche and species formation in genetic function optimization. Proceedings of the Third International Conference on Genetic Algorithms (pp. 42--50).

....In this section we try to find the two optima of respectively a ### bit TwoMax, and a ### bit Ising model using an srGA with this sharing method. To avoid linkage problems on the Ising model, the algorithm uses two points crossover. As sharing function the triangular sharing function defined in [1] is used. The sharing threshold is calculated by # ## # # ## # # ### with # the problem s number of optima. Individuals at Hamming distance smaller than #### share their fitness. When optimizing an Ising model, equally good partial solutions should be kept in the population. Experiments show ....

K. Deb and D. E. Goldberg. An investigation of niche and species formation in genetic function optimization. In J. D. Schaffer, editor, Proceedings of the 3rd International Conference on Genetic Algorithms, pages 42--50. Morgan Kaufmann, 1989.

....metric chosen to calculate the proximity measure between two population members. The parameter ashare denotes the largest value of that distance metric within which any two solutions share each other s fitness. This parameter is usually set by the user, although there exist some guidelines [3]. There are two difficulties with this sharing function approach: 1. The performance of the sharing function method in maintaining a spread of solutions largely depends on the chosen (Yshare value. 2. Since each solution must be compared with all other solutions in the population, the overall ....

K. Deb and D. E. Goldberg, "An investigation of niche and species formation in genetic function optimization,", in Proceedings of the Third International Conference on Genetic Algorithms, J. D. Schaffer, Ed. San Mateo, CA: Morgan Kauffman, 1989, pp. 42-50.

No context found.

Deb, K., Goldberg, D. E.: An Investigation of Niche and Species Formation in Genetic Function Optimization. In: Schaffer, J. D. (ed.): Proceedings of Third International Conference of Genetic Algorithms, Morgan Kaufmann, San Mateo, CA (1989) 42-50

No context found.

K. Deb and D. E. Goldberg, "An investigation of niche and species formation in genetic function optimization," in Proc. 3rd Int. Conf. Genetic Algorithms, J. D. Schaffer, Ed., Washington, DC, 1989, pp. 42--50.

No context found.

Deb, K. and Goldberg, D. (1989). An investigation of niche and species formation in genetic function optimization. In Schaffer, J., editor, Proceedings of the Third International Conference on Genetic Algorithms, pages 42--50, San Mateo, CA, USA. Morgan Kaufmann.

No context found.

K. Deb and D. E. Goldberg. An Investigation of Niche and Species Formation in Genetic Function Optimization. In J. D. Schaffer, editor, Proceedings of the Third International Conference on Genetic Algorithms, pages 42--50, San Mateo, California, June 1989. George Mason University, Morgan Kaufmann Publishers.

No context found.

K. Deb and D. E. Goldberg. An investigation of niche and species formation in genetic function. In Proc. of 3 Algorithms, pages 4250, 1989.

No context found.

K. Deb and D. E. Goldberg, "An investigation of niche and species formation in genetic function," Proc. 3rd Int. Conf. Genetic Algorithms, 1989, pp. 42--50.

No context found.

K. Deb and D. E. Goldberg. An investigation of niche and species formation in genetic function. In Proc. of International Conference on Genetic Algorithms, pages 4250, 1989.

No context found.

Kalyanmoy Deb and David E. Goldberg. An Investigation of Niche and Species Formation in Genetic Function Optimization. In J. David Schaffer, editor, Proceedings of the Third International Conference on Genetic Algorithms, pages 42-- 50, San Mateo, California, June 1989. George Mason University, Morgan Kaufmann Publishers.

No context found.

K. Deb and D. E. Goldberg. An investigation of niche and species formation in genetic function. In Proc. of 3 Algorithms, pages 4250, 1989.

No context found.

Deb, K., Goldberg, D.E.: An investigation of niche and species formation in genetic function optimization. In: Proc. ICGA. (1989) 42--50

No context found.

Deb K and Goldberg D E . An Investigation of Niche and Species Formation in Genetic Function Optimization Proceedings of The Fifth International Conference on Genetic Algorithms and Their Applications. Morgan Kaufmann,1989, 42-50.

First 50 documents  Next 50

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC