Kargupta, H., and R. E. Smith. 1991. System identification with evolving polynomial networks. In Proceedings of the fourth international conference on genetic algorithms, 370376.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....sort of implicit niching into the GA. In many studies, the poor performance of the genetic system could be attributed to a poor method of niching. Fitness sharing (Goldberg Richardson, 1987) is one type of niching method that has been applied to classification (Booker, 1982; Horn et al. 1994; Kargupta Smith, 1991; Packard, 1990; Smith Valenzuela Rendn, 1989) Fitness sharing works by reducing the fitnesses of similar population elements. Crowding (De Jong, 1975; Mahfoud, 1992, 1994, 1995a) is another type of niching method that has also been applied to classification (Booker, 1982; Goldberg, 1983; ....

Kargupta, H., and R. E. Smith. 1991. System identification with evolving polynomial networks. In Proceedings of the fourth international conference on genetic algorithms, 370376.

....like back propagation to find appropriate connection weights [MD89, TSVdM93, YHLK94] ii) evolutionary computation has been used to determine good network connection schemes [MTH89, Kit90, Man93] There are also studies that have a closer relation to our approach. Examples are the studies of [KS91] and [Rog91] Both papers concern evolutionary algorithms to search for functions to be used for input transformation. However, different basic functions, and different networks are used in in these studies. The remainder of this paper is structured as follows: in section 2 a more precise ....

H. Kargupta and R.E. Smith. System identification with evolving polynomial networks. In Proceedings of the Fourth International Conference on Genetic Algorithms, pages 370--376. Morgan Kaufmann, 1991.

....of a tree is done by replacing subtrees by random subtrees. Since inconsistent expressions may originate some mathematical operators are protected so that, if an inconsistency occurs, a zero is returned to prevent the algorithm to stuck. 4 Related methods Methods that are related are for example [KS91] and [Rog91] Algorithms more or less related are, amongst others: multilayer networks; neural networks of which the architecture is determined using genetic algorithms; and genetic programming. For an extensive overview of combinations of neural networks and genetic algorithms see [Bra95] ....

H. Kargupta and R.E. Smith. System identification with evolving polynomial networks. In Proceedings of the Fourth International Conference on Genetic Algorithms, pages 370--376. Morgan Kaufmann, 1991.

....for solving complex optimization problems whose variables vary widely in their sensitivity to the objective function. I. Introduction Genetic algorithms (GAs) have been demonstrated to be a promising search and optimization technique [1] It has been successfully applied to system identification [2, 3, 4, 5] and a wide range of applications including design [6] scheduling [7] routing [8] control [9, 10] and others [11, 12, 13] One of the main obstacles in applying GAs to complex problems has often been the high computational cost due to their slow convergence rate. The convergence rate of a GA ....

H. Kargupta and R. E. Smith, "System identification with evolving polynomial networks," In Proceedings of the Fourth International Conference on Genetic Algorithms, San Diego, CA, July 1991.

....like back propagation to find appropriate connection weights [MD89, TSVdM93, YHLK94] ii) evolutionary computation has been used to determine good network connection schemes [MTH89, Kit90, Man93] There are also studies that have a closer relation to our approach. Examples are the studies of [KS91] and [Rog91] Both papers concern evolutionary algorithms to search for functions to be used for input transformation. However, different basic functions, and different networks are used in in these studies. The remainder of this paper is structured as follows: in Section 2 a more precise ....

H. Kargupta and R.E. Smith. System identification with evolving polynomial networks. In Proceedings of the Fourth International Conference on Genetic Algorithms, pages 370--376. Morgan Kaufmann, 1991.

....suitable for solving optimization problems whose variables vary widely in their sensitivity to the objective function. I. Introduction Genetic algorithms (GAs) have been demonstrated to be a promising search and optimization technique [1] It has been successfully applied to system identification [2, 3, 4, 5] and a wide range of applications including design [6] scheduling [7] routing [8] control [9, 10] and others [11, 12, 13] One of the main obstacles in applying GAs to complex problems has often been the high computational cost due to their slow convergence rate. The convergence rate of a GA ....

H. Kargupta and R. E. Smith, "System identification with evolving polynomial networks," In Proceedings of the Fourth International Conference on Genetic Algorithms, San Diego, CA, July 1991.

....these systems are similar to those in the immune system simulations with oe = N . The techniques presented here may also prove useful in other systems that form computational networks via genetic learning. A particular instance is in the development of poly nomial networks for system modeling (Kargupta Smith, 1991), where explicit fitness sharing was previously required. Whether or not the techniques used in the immune system simulations can be explicitly transferred to more prescriptive applications, analysis of the type used in this paper can aid in understanding GA behavior in settings that require ....

Kargupta, H., & Smith, R. E. (1991). System identification with evolving polynomial networks. In Proceedings of the Forth International Conference on Genetic Algorithms, 370--376, San Mateo, CA. Morgan-Kaufmann.

....the emergent effects in these systems are similar to those in the immune system simulations. The techniques may also prove useful in other systems that form computational networks via genetic learning. A particular instance is in the development of polynomial networks for system modeling (Kargupta Smith, 1991), where explicit fitness sharing was previously required. Whether or not the techniques used in the immune system simulations can be explicitly transferred to more prescriptive applications, analysis of the type used in this paper can aid in understanding GA behavior in settings that require ....

Kargupta, H., & Smith, R. E. (1991). System identification with evolving polynomial networks (TCGA Report No. 91001). Tuscaloosa: University of Alabama, The Clearinghouse for Genetic Algorithms.

No context found.

H. Kargupta and R. E. Smith, "System identification with evolving polynomial networks," in Proc. Fourth Int. Conf. Genetic Algorithms (ICGA-91), R. Belew and L. Booker, Ed. Morgan Kaufmann, 1991, pp. 370--376.

No context found.

Hillol Kargupta and Robert E. Smith. System Identification with Evolving Polynomial Networks. In Proceedings of the Fourth International Conference on Genetic Algorithms, pages 370--376, 1991.

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