Fonseca, C. M. and Fleming, P. J. (1996). Nonlinear System Identification with Multiobjective Genetic Algorithms. In Proceedings of the 13th World Congress of the International Federation of Automatic Control, pages 187--192.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....parameter selections found is given in Table 6. The results presented in Figure 4 and Table 5 are for these settings. 4 Results 4. 1 Performance Metrics As in previous research, we measure the performance of the algorithms tested using a statistical comparative assessment technique adapted from [5]. We refer the reader to [14, 16] for a complete description of our implementation of the technique and a discussion of its advantages and disadvantages. Here, the reader need only understand that the technique allows us to present results in two ways. First, as a pair of numbers [a; b] ....

C. M. Fonseca and P. J. Fleming. Nonlinear System Identification with Multiobjective Genetic Algorithms. In Proceedings of the 13th World Congress of the International Federation of Automatic Control, pages 187-- 192, San Francisco, California, 1996. Pergamon Press.

....do not improve the solution. This is similar to the method used in M PAES. Once again, l fails was set empirically. 4 Results 4. 1 Performance Metrics As in previous research, we measure the performance of the algorithms tested, using a statistical comparative assessment technique adapted from [4]. We refer the reader to [8, 11] for a complete description of our implementation of the technique and a discussion of its advantages and disadvantages. objective 1 objective 2 x o x x o x o x o x Figure 3: A collection of two sets of nondominated vectors. Each set of vectors defines ....

C. M. Fonseca and P. J. Fleming. Nonlinear System Identification with Multiobjective Genetic Algorithms. In Proceedings of the 13th World Congress of the International Federation of Automatic Control, pages 187--192, San Francisco, California, 1996. Pergamon Press.

....sharing in an attempt to find a uniform (equidistant) distribution of vectors representing PF true , i.e. one in which PF known s shape is a good Evolutionary Computation Volume 8, Number 2 137 D. Van Veldhuizen and G. Lamont 2 0 2 2 0 2 0 100 200 300 x value Fitness vs. Genotype (Fonseca) a) y value Pareto Ranking 2 0 2 2 0 2 0 10 20 x value Fitness vs. Genotype (Goldberg) b) y value Pareto Ranking 2 0 2 2 0 2 0 0.5 1 x value Fitness vs. Genotype (Simple) c) y value Pareto Ranking Figure 4: Pareto ranking schemes. approximation of PF true . We compare selected implementations ....

Fonseca, C. M. and Fleming, P. J. (1997b). Non-Linear System Identification with Multiobjective Genetic Algorithms. In Proceedings of the Thirteenth Triennial World Congress of the International Federation of Automatic Control, pages 187--192, Pergamon, London, England.

....parameter selections found is given in Table 6. The results presented in Figure 4 and Table 5 are for these settings. 4 Results 4. 1 Performance Metrics As in previous research, we measure the performance of the algorithms tested using a statistical comparative assessment technique adapted from [5]. We refer the reader to [15, 17] for a complete description of our implementation of the technique and a discussion of its advantages and disadvantages. Here, the reader need only understand that the technique allows us to present results in two ways. First, as a pair of numbers [a,b] indicating ....

C. M. Fonseca and P. J. Fleming. Nonlinear System Identification with Multiobjective Genetic Algorithms. In Proceedings of the 13th World Congress of the International Federation of Automatic Control, pages 187-- 192, San Francisco, California, 1996. Pergamon Press.

....sizes and specialized operators. MOGP was used for the identification of non linear model structures, as an alternative that the authors reported to work better (in terms of representation power) than the use of the conventional linear representation of MOGA that they had attempted before [Fonseca and Fleming 1996a] Aherne et al. 1997] used MOGA to optimize the selection of parameters for an object recognition scheme called the Pairwise Geometric Histogram paradigm. Todd and Sen [1997] used a variant of MOGA for the preplanning of containership layouts (a large scale combinatorial problem) In Todd and ....

....is to determine how to measure the quality of a solution. So far, practically visual inspection is the only technique used, unless there is some previous knowledge of the points which lie in the Pareto front (in which case there is obviously no need for a multiobjective optimization technique) Fonseca and Fleming [1996b] proposed the definition of certain (arbitrary) goals that we wish the GA to attain; then we can perform multiple runs and apply standard non parametric statistical procedures to evaluate the quality of the solutions (i.e. the non dominated fronts) An Updated Survey of GA Based Multiobjective ....

Fonseca, C. M. and Fleming, P. J. 1996a. Nonlinear System Identification with Multiobjective Genetic Algorithms. In Proceedings of the 13th World Congress of IFAC (San Francisco, California, 1996), pp. 187--192.

....sizes and specialized operators. MOGP was used for the identification of non linear model structures, as an alternative that the authors reported to work better (in terms of representation power) than the use of the conventional linear representation of MOGA that they had attempted before [33]. Aherne et al. 34] used MOGA to optimize the selection of parameters for an object recognition scheme called the Pairwise Geometric Histogram paradigm. Todd and Sen [35] used a variant of MOGA for the preplanning of containership layouts (a large scale combinatorial problem) In Todd and Sen s ....

Carlos M. Fonseca and Peter J. Fleming. Nonlinear System Identification with Multiobjective Genetic Algorithms. In Proceedings of the 13th World Congress of the International Federation of Automatic Control, pages 187--192, San Francisco, California, 1996. Pergamon Press.

....sizes and specialized operators. MOGP was used for the identification of non linear model structures, as an alternative that the authors reported to work better (in terms of representation power) than the use of the conventional linear representation of MOGA that they had attempted before [22]. Aherne et al. 1] used MOGA to optimize the selection of parameters for an object recognition scheme called the Pairwise Geometric Histogram paradigm. Todd and Sen [94] used a variant of MOGA for the preplanning of containership layouts (a large scale combinatorial problem) In Todd and ....

Carlos M. Fonseca and Peter J. Fleming. Nonlinear System Identification with Multiobjective Genetic Algorithms. In Proceedings of the 13th World Congress of IFAC, pages 187--192, San Francisco, California, 1996.

....the problem of identifying non linear model structures cannot be formulated using only a single criterion. Identification involves diverse characteristics as linearity, degree of non linearity, model structure, performance and model validation, which have to be considered. In earlier work (Fonseca and Fleming, 1996), genetic algorithms have been applied to identify non linear model structures where system identification has been formulated as a subset selection problem. More recently, Rodrguez et al. 1997) have used a tree structured representation (Koza, 1992) to this problem. This approach is based on an ....

....also provides the opportunity of manipulating the family of solutions by changing goals values of the objective functions depending on the purpose of the identification procedure. While the framework has previously been applied to identification problems using the subset representation approach (Fonseca and Fleming, 1996), the hierarchical tree encoding used here appears to be more powerful. Because, in subset selection, the total set of all possible linear and non linear terms and the maximum length of the chromosome, corresponding to the maximum number of terms, have to be set up a priori, the richness of the ....

Fonseca, C.M. and P.J. Fleming. (1996) Nonlinear System Identification with Multiobjective Genetic Algorithms, Proc. of the 13th World Congress of IFAC, San. Fco., 187-192.

No context found.

Fonseca, C. M. and Fleming, P. J. (1996). Nonlinear System Identification with Multiobjective Genetic Algorithms. In Proceedings of the 13th World Congress of the International Federation of Automatic Control, pages 187--192.

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