Rudnick, M. (1990). A bibliography of the intersection of genetic search and artificial neural networks. Techn. Rep. No. CS/E 90-- 001. Oregon Graduate Center, Beaverton, OR.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....for a specific application is still very much a research issue and since, from the evolutionary viewpoint, Nature has performed this task magnificently it is natural to ask to what extent we might use our present knowledge of GAs to design neural networks. Bibliographies of work in this field [Rudnick 1990] and [Weiss 1990] are available up to 1990. In such a paradigm we might expect the computational cost to be high but it is possible that from a number of such studies we may begin to extract our own design rules by examining the solutions produced by the GA. A very early example of such work is ....

M. Rudnick. A bibliography of the intersection of genetic search and neural networks. Technical Report CS/E 90-001, Oregeon Graduate Centre, 1990.

....nodes. The paper closes with a summary and discussion of the results presented, and draws out some of the wider implications for genetic search in other domains. 2 Genetic Approaches to Neural Networks Genetic algorithms are increasingly being applied to problems in neural networks. Rudnick [24] and Weiss [34] have produced excellent bibliographies for this field in 1990. A number of approaches can be distinguished, all of which have had limited success, and most of which have concentrated on Rumelhart type feed forward networks. The two primary areas of activity have been: 1. Topology ....

Mike Rudnick. A Bibliography of the Intersection of Genetic Search and Neural Networks. Technical Report CS/E 90-001, Oregon Graduate Centre, 1990.

....and the fitness evaluation consists of constructing the network, testing the network, and obtaining a fitness assessment based on the performance of the network. Examples are provided in the overviews by Yao [80] Wei [75] and Schaffer et al. 67] A useful early bibliography is given by Rudnick [63]) The approaches taken can be broadly categorised as evolving fixed architecture neural networks, and evolving the network architecture itself. In both cases the activation function of each neuron is the same. These two approaches are described in the following sections. 3.2.1 Fixed ....

Mike Rudnick. A bibliography of the intersection of genetic search and artificial neural networks. Technical Report CS/E 90-001, Department of Computer Science and Engineering, Oregon Graduate Institute, January 1990.

....of each technique lends credence to the emulation of nature for purposes of constructing adaptive and robust artifacts. With the GA and ANN models in place, an obvious step is to combine the two in order to evolve ANNs. Indeed, much research has proceeded along these lines. Refer to [44] 49] [39], and [6] for extensive review and references. 27 Evolving Connection Strengths Supervised learning in an ANN is usually formulated as a procedure for searching for network connection strengths (weights) that meet an optimality measure. For instance, a popular learning algorithm for acyclic ANNs ....

Mike Rudnick. A bibliography of the intersection of genetic search and artificial neural networks. Technical Report CS/E 90-001, Deparment of Computer Science and Engineering, Oregon Graduate Institute, January 1990.

....possible architectures for the more appropriate ones for the task in hand. Approaches to evolving connectionist network architectures that have been taken include those in (Kerszberg and Bergman 1988, Harp et al. 1989, Miller et al. 1989, Muhlenbein and Kindermann 1989) see also references in (Rudnick 1990). The genotype must specify the characteristics of nodes, and the connections between nodes. This can be done in either a serial descriptive fashion, or through some form of developmental process as will be discussed in Chapter 10. The number of different ways of doing this, such as those in the ....

M. Rudnick. A bibliography of the intersection of genetic search and artificial neural networks. Technical Report CS/E 90-001, Oregon Graduate Institute, Dept. of Computer Science and Engineering, 1990.

....Dave Opitz, Larry Yaeger, Martin Mandischer , Jari Vaario, Russell Anderson, Riccardo Poli, Peter Hancock, Javier Marin, Frank Amos, Egbert Boers, Spyros Kazarlis and Johannes Sch afer for sending us relevant bibliographic references. We have also drawn on earlier bibliographies on this topic [Rudnick, 1990, Schaffer, 1994] We are grateful to National Science Foundation for its support of our research through a grant (IRI 9409580) to Vasant Honavar. 1 GANN: Genetic Algorithms and Neural Networks Mailing List. To subscribe, send e mail to gann request cs.iastate.edu with subject subscribe. ....

....1992] Michalewicz, Z. Genetic Algorithms Data Structures = Evolution Programs. New York, NY: Springer Verlag. Miller et al., 1989] Miller, G. Todd, P. Hegde, S. Designing Neural Networks Using Genetic Algorithms. Proceedings of the Third International Conference on Genetic Algorithms 379 384. [Rudnick, 1990] Rudnick, M. A Bibliography: The Intersection of Genetic Search and Artificial Neural Networks. Technical Report CSE 90 001, Department of Computer Science and Engineering, Oregon Graduate Institute. Rumelhart McClelland, 1986] Rumelhart, D. McClelland, J. Ed. Parallel Distributed ....

Michael Rudnick. A Bibliography: The Intersection of Genetic Search and Artificial Neural Networks. Cs/e 90-001, Department of Computer Science and Engineering, Oregon Graduate Institute, 1990.

....nodes. The paper closes with a summary and discussion of the results presented, and draws out some of the wider implications for genetic search in other domains. 2 Genetic Approaches to Neural Networks Genetic algorithms are increasingly being applied to problems in neural networks. Rudnick [24] and Weiss [34] have produced excellent bibliographies for this field in 1990. A number of approaches can be distinguished, all of which have had limited success, and most of which have concentrated on Rumelhart type feed forward networks. The two primary areas of activity have been: 1. Topology ....

Mike Rudnick. A Bibliography of the Intersection of Genetic Search and Neural Networks. Technical Report CS/E 90-001, Oregon Graduate Centre, 1990.

....The focus lies on problem specific aspects, not so much on the applications or test problems used or GA design features that do not directly relate to the neural network problem. Also, the article does not claim to be complete. For supplementary overviews on the subject, the reader is referred to [61, 65, 74] or [7] taxonomy and guide to literature) A bibliography with 510 related articles can be found in [2] In most cases, Evolutionary Algorithms have been used to either train the network or to find a suitable topology. This is reflected in the outline of the paper: Section 2 is concerned with ....

M. Rudnick. A bibliography of the intersection of genetic search and artificial neural networks. Technical report, Oregon Graduate Institute, Department of Computer Science and Engineering, January 1990.

....rule, input feature selection, genetic reinforcement learning, initial weight selection, ANN analysis, etc. Almost all ANN conferences have papers on such combinations. There is even a workshop solely devoted to this area [1] The emphasis of this paper is rather different from other review papers [2, 3, 4, 5]. This paper is more concerned with the understanding and development of EANNs, which can be regarded as a general framework for adaptive systems, i.e. systems that can change their architectures and learning rules according to different environments without human intervention. Other ....

M. Rudnick. A bibliography of the intersection of genetic search and artificial neural networks. Technical Report CS/E 90-001, Department of Computer Science and Engineering, Oregon Graduate Institute of Science and Technology, January 1990.

....speaking, evolutionary methods represent a very general technique because they can be applied to any aspect of the agent (synaptic connections and number of neurons [46] growing factors [6] body specifications [29] etc. and to any type of algorithm instantiation (neural networks [42], classifier systems [12] graph systems [26] programming language functions [27] etc. as long as they can be mapped into a genetic description. It has been empirically shown that genetic algorithms outperform other search methods when the space of the possible solutions is high dimensional ....

M. Rudnick. A Bibliography of the Intersection of Genetic Search and Artificial Neural Networks. Technical Report CS/E 90-001, Department of Computer Science and Engineering, Oregon Graduate Center, January 1990.

....three kinds of evolution, but also considers interactions among them. Keywords Artificial Neural Networks, Learning, Evolution, Genetic Algorithms. X. Yao: A Review of Evolutionary Artificial Neural Networks 1 1 Introduction The interest in EANNs has been growing rapidly in recent years [1, 2, 3, 4], as the research can not only further our understanding of adaptive processes in nature, but also help computer scientists and engineers develop more powerful artificial systems. This paper mainly serves the second purpose. Since we are most interested in exploring possible benefits arising from ....

M. Rudnick. A bibliography of the intersection of genetic search and artificial neural networks. Technical Report CS/E 90-001, Department of Computer Science and Engineering, Oregon Graduate Institute of Science and Technology, January 1990.

....suited to the requirements of the task, it is still premature to assess the superiority of one approach over the other for the evolution of real robots. VI. Related work There is a large literature on the application of evolutionary techniques to the design and training of neural networks (see [47] for a specific bibliography, 48] 49] 50] for a description of the various approaches employed, and the 1993 special issue on Evolutionary Computation of the IEEE Transactions on Neural Networks for an outline of more recent results) Only a small subset of this corpus of research focuses on ....

M. Rudnick, "A Bibliography of the Intersection of Genetic Search and Artificial Neural Networks", Tech. Rep. CS/E 90001, Department of Computer Science and Engineering, Oregon Graduate Center, January 1990.

....evolution and ANNs together [Dew85, Dre87, DrK87, Dre89, Dre90a, Dre90b, Joh88, Lan89, Rin86, Rin] 2.7. Miscellaneous Here are several works that don t seem to fit neatly into any of the categories I ve arranged [Ack87, CaD89, CaD90, DoD87, KoH88, SoS88] I also include this bibliography here [Rud90]. 2.8. Other Finally, there are several works that I ve not attempted to categorize because I ve not read them, or perhaps have briefly looked at them but wasn t able to see a ready category [BeK87, Cas89, Fen86, HaW84, HaW86, Oos89, WaH85] 3. Neuro Evolution Mailing List I operate a mailing ....

M. Rudnick (1990). "A Bibliography: The Intersection of Genetic Search and Artificial Neural Networks," CS/E 90-001 , Department of Computer Science and Engineering, Oregon Graduate Institute.

No context found.

Rudnick, M. (1990). A bibliography of the intersection of genetic search and artificial neural networks. Techn. Rep. No. CS/E 90-- 001. Oregon Graduate Center, Beaverton, OR.

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