Robertson, G. G., 1987, Parallel implementation of genetic algorithms in a classifier system, in: Proc. 2nd Int. Conf. on Genetic Algorithms, pp. 140-- 147.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....small demes. The demes overlap providing a way to disemminate good solutions across the entire population. Again, selection and mating occur only within a deme. In the second ICGA, Robertson published a paper describing the parallelization of a GA in a classifier system using a Connection Machine [55]. He parallelized the selection of parents and of classifiers to replace and also the mating and crossover operations. The execution time of this algorithm is independent of the number of 26 classifiers (up to 16K) In 1989 the ASPARAGOS system was introduced in the paper by Gorges Schleuter ....

George G. Robertson. Parallel Implementation of Genetic Algorithms in a Classifier System. In John J. Grefenstette, editor, Proceedings of the 2nd International Conference on Genetic Algorithms (ICGA87), pages 140--147, Cambridge, MA, July 1987. Lawrence Erlbaum Associates. Also Technical Report TR-159 RL87-5 Thinking Machines Corporation.

....of the global population model. The more difficult problem, to bring the mating individuals together without memory conflicts or communication bottlenecks, remains. 3 3. 2 Implementations of Global Population Models The first complete parallel implementation of the global model was done by ROBERTSON (1987) on a CM 1 with 65536 processors. This array processor provides several parallel operators, that help parallelizing the selection operators. In Figure 2 some important implementations of global population models are summarized. Reference Computer (PEs) Selection Reproduction Evaluation ROBERTSON ....

....(1987) on a CM 1 with 65536 processors. This array processor provides several parallel operators, that help parallelizing the selection operators. In Figure 2 some important implementations of global population models are summarized. Reference Computer (PEs) Selection Reproduction Evaluation ROBERTSON (1987) TMC CM 1 (65536) parallel parallel parallel FOGARTY et al. 1991) T800 Network (4 72) sequential sequential parallel ABRAMSON et al. 1992) EncoreMultimax (15) sequential parallel parallel (1993) Fujitsu AP1000 (128) sequential parallel parallel PUNCH et al. 1993) BBN GP1000 (51) sequential ....

ROBERTSON, G. G. 1987. Parallel Implementation of Genetic Algorithms in a Classifier System. In: (GREFENSTETTE, 1987), 140--147.

....W. 117] Reorda, M. S. 262] Resende, M.G. C. 187] Ribeiro Filho, J. 126] Ribeiro Filho, Jose L. 118] Richards, Dana S. 374, 376, 377, 378, 379] Richards, W. 162] Richert, P. 486] Richter, K. R. 117] Riessen, G. A. 327] Riveros, J. Fernando V. 328] Robertson, George G. [479] Roda, J. 534] Rodriguez, C. 534] Romero R. Milton E. 328] Ronde, J. F. de, 265, 329] Rosati, M. 259, 313, 325] Rosato, V. 259, 313, 325] Roupec, Jan, 185] Rowe, Jon, 243, 263] Roychowdhury, Vwani P. 68] Roysam, Badrinath, 554, 555] Rubin, S. 66] Rudnick, Elizabeth M. ....

.... 38, 39] cellular GA, 190] chaos, 45] chemistry macromolecule, 321] physical, 95, 259, 313] reaction kinetics, 55] RNA secondary structure, 83] structural, 325, 342, 347] chromosomes xing, 154] classi er implementation ALECSYS, 386, 387] classi er systems, 464, 81, 183] classi ers, [479, 386, 387, 412, 430, 110] clustering, 413, 106, 139, 216, 246] constrained, 322] fuzzy, 273] clusters metal, 325] coding real, 338, 347] comparison parallel methods in TSP, 467] comparison Breeder GA, 240] evolution strategies, 323] other optimization methods, 347] parallel GA, 472, 245, 272] scheduling ....

[Article contains additional citation context not shown here]

George G. Robertson. Parallel implementation of genetic algorithms in classier systems. In Lawrence Davis, editor, Genetic Algorithms and Simulated Annealing, pages 129-140. Pitman Publishing, London, 1987. ga:Robertson87a.

.... applications that can be parallelized efficiently (such as genetic algorithms, genetic programming, and other techniques of evolutionary computation) Amenability to parallelization is a recognized feature of genetic algorithms, genetic programming, and other evolutionary algorithms (Holland 1975; Robertson 1987; Tanese 1989; Goldberg 1989; Stender 1993; Koza and Andre 1995; Andre and Koza 1996a, 1996b) Section 2 describes one commonly used approach to parallelization of evolutionary algorithms, namely the asynchronous island approach involving semi isolated subpopulations. Section 3 points out that a half peta flop of computational power is ....

Robertson, George. l987. Parallel implementation of genetic algorithms in a classifier system. In Davis, Lawrence. (editor). Genetic Algorithms and Simulated Annealing London: Pittman. Stender, Joachim (editor). 1993. Parallel Genetic Algorithms. Amsterdam: IOS Publishing.

....value, then the child may still be saved by virtue of the Boltzman distribution, with temperature T) Some variants of this algorithm have also been implemented. It is the iterative nature of this genetic algorithm that permits it to be parallelized. Similar work has been done by Ackley (1987) and Robertson (1987). 8 Conclusion Prolog D Linda is a truly distributed logic programming environment. The distribution allows applications to take advantage of the added computing power available, as well as to be structured in a parallel fashion. The parallelism obtained is acknowledged to be coarse 3 . In the ....

Robertson G. (1987), Parallel Implementation of Genetic Algorithms in a Classifier System, In Davis L. (Ed.), Genetic Algorithms and Simulated Annealing, (Research Notes in Artificial Intelligence), Pitman Publishing, London, England, 129-140.

....parallel GAs have only one population, but it is has a spatial structure that limits the interactions between individuals. An individual can only compete and mate with its neighbors, but since the neighborhoods overlap good solutions may disseminate across the entire population. Robertson [ROB 87] parallelized the genetic algorithm of a classifier system on a Connection Machine 1. He parallelized the selection of parents, the selection of classifiers to replace, mating, and crossover. The execution time of his implementation was independent of the number of classifiers (up to 16K, the ....

ROBERTSON G. G., Parallel implementation of genetic algorithms in a classifier system ». In GREFENSTETTE J. J., Ed., Proceedings of the Second International Conference on Genetic Algorithms, p. 140--147, Lawrence Erlbaum Associates (Hillsdale, NJ), 1987.

....strength. Actually the CFS C system also provides for several types of taxes , e.g. a tax on every classifier every step, a tax for bidding, and so on, which are described later in this chapter. I have ignored those in this derivation of the fixed point strength of a classifier. 5 Robertson [Robertson, 1987] has shown that almost all aspects of learning classifier systems are parallelizable by implementing a classifier system on a Connection Machine. In order to alleviate that problem, I have introduced a new variable, DMShare (detector message share) into the CFS C system. The basic idea is to ....

Robertson, George G. "Parallel Implementation of Genetic Algorithms in a Classifier System." In Proceedings of the Second International Conference on Genetic Algorithms and their Applications, 140-147. John J. Grefenstette (Ed.). Cambridge, Massachesetts, July 28-31, 1987.

....The cross over operation. We use both operators to produce new elements : first we use the cross over and then the mutation on the newly produced element. 5.1. 2 Principle of the parallel genetic algorithm (PGA) There are many parallel versions of genetic algorithms : the parallel standard version [31], the decomposition version [32] 33] and the massively parallel version [29] We chose this last method. The principle is to place one element of the population per processor so the steps 1, 3 and 4 can be executed in parallel. Furthermore, the selection (step 2) is made locally, each individual ....

G.Robertson, Parallel implementation of genetic algorithms in a classifier system, in Genetic algorithms an simulated annealing, L.Davis (editor), Morgan Kaufmann Pub., 1987.

....any global control structure. Selection is performed, cell by cell, only over the individual assigned to that cell and its neighbours; crossover mates neighbouring individuals, while mutation is standard. In addition, other operators have been proposed, which will be outlined in the following. Robertson (1987) parallelised the genetic algorithm of a classifier system on a Connection Machine 1. He parallelised selection of parents, the selection of classifiers to replace, mating, and crossover. The execution time of his implementation was independent of the number of classifiers (up to 16K, the number ....

Robertson, G. (1987). Parallel implementation of genetic algorithms in a classifier system, in L. Davis (ed.), Genetic Algorithms and Simulated Annealing, Pittman, London.

....for parallelisation (Goldberg 1989) Tanese 1989a) because their time consuming fitness evaluations can be performed independently for each individual in the population. Existing work on parallel GAs include the global, coarsegrained and fine grained models (C.H. Pettey and Grefenstette 1987) (Robertson 1987), Cantu Paz and Mej iaOkvera n.d. Grosso 1985) Spiessens and Manderick 1990) Gorges Schleuter 1991) Gorges Schleuter 1992) Parallel GPs have been implemented on a network of transputers (Andre and Koza 1995) We propose a Java based distributed approach to parallelise GPs on the ....

Robertson, G. (1987). Parallel implementation of genetic algorithms in a classifier system, in L. Davis (ed.), Genetic Algorithms and Simulated Annealing, Pittman, London.

....made them very popular in the very last years. They have recently been applied to combinatorial optimization problems in various fields, such as, for instance, the traveling salesman problem [4] the optimization of connections and connectivity of neural networks [5] and classifier systems [6]. Standard genetic algorithms with large populations take an extremely long time to execute. We have therefore proposed a parallel algorithm to speed up the genetic process. The purpose of this paper is to prove that the mapping problem may be solved quite efficiently by a parallel genetic ....

G.Robertson, "Parallel implementation of genetic algorithms in a classifier system", in Genetic algorithms and Simulated annealing, L.Davis ed., Morgan Kaufmann Publishers, pp.129-140, 1987.

....value, then the child may still be saved by virtue of the Boltzman distribution, with temperature T. Some variants of this algorithm have also been implemented. It is the iterative nature of this genetic algorithm that permits it to be parallelised. Similar work has been done by Ackley#[1987] and Robertson [1987]. 8 Conclusion Prolog D Linda is a truly distributed logic programming environment. The distribution allows applications to take advantage of the added computing power available, as well as to be structured in a parallel fashion. The parallelism obtained is acknowledged to be coarse. In the ....

Robertson G. (1987), Parallel Implementation of Genetic Algorithms in a Classifier System, In Davis L. (Ed.), Genetic Algorithms and Simulated Annealing, (Research Notes in Artificial Intelligence), Pitman Publishing, London, England, 129-140.

....execution models of GAs have been proposed and applied to solve various problems. Three approaches to parallel genetic algorithms may be considered [Tal95] ffl the standard parallel approach: in this approach, the evaluation, the selection and the reproduction steps are performed in parallel [Rob87] Both mating and selection are performed over the whole population. ffl the decomposition approach: this approach consists in dividing the population into equally sized subpopulations. Each processor runs the genetic algorithm on one subpopulation, periodically selecting good individuals to ....

G. Robertson. Parallel implementation of genetic algorithms in a classifier system. Genetic algorithms and simulated annealing, L.Davis ed., Morgan Kaufmann Pub., pages 129--140, 1987.

....only exploiting the power of parallel computers. We have therefore developed a parallel distributed system that can be used as a tool to build genetics based machine learning GBML systems. A parallel implementation of a GBML system on the Connection Machine has been proposed by Robertson [1]. That work demonstrates the power of such a solution, but still retains, we think, a basic limitation: as the Connection Machine is a SIMD architecture the resulting implementation is only a more powerful but still classic GBML system. To implement our system we have used a transputer net that, ....

Robertson,G.G., "Parallel Implementation of Genetic Algorithms in a Classifier System", Proceedings of the Second International Conference on Genetic Algorithms, July 28-31 1987, Lawrence Erlbaum. 12

No context found.

Robertson, G. G., 1987, Parallel implementation of genetic algorithms in a classifier system, in: Proc. 2nd Int. Conf. on Genetic Algorithms, pp. 140-- 147.

No context found.

G. Robertson, "Parallel implementation of genetic algorithms in a classifier system", Genetic Algorithms and Simulated Annealing, London: Pittman, 1987.

No context found.

Robertson, G. G. (1987). Parallel implementation of Genetic Algorithms in a Classifier System. In Proceedings of the 2nd International Conference on Genetic Algorithms, pages 155--161. Lawerence Erlbaum Associates Publishers.

No context found.

Robertson, G. G. (1987). Parallel implementation of genetic algorithms in a classier system. Proceedings of the Second International Conference on Genetic Algorithms,

No context found.

George G. Robertson. Parallel Implementation of Genetic Algorithms in a Classier System. In Davis [104], pages 129-140.

No context found.

George G. Robertson. Parallel Implementation of Genetic Algorithms in a Classier System. In Grefenstette

No context found.

George Robertson, Parallel Implementations of Genetic Algorithms in a Classifier System, Cambridge, Mass., Thinking Machines Corporation, Technical Report Series RL87-5.

No context found.

Robertson, G. "Parallel Implementation of Genetic Algorithms in a Classifier Design".Research Notes on Artificial Intelligence, Genetic Algorithms and Simulated Annealing, Pittman, 1987, London.

No context found.

Robertson, George G. (1987). Parallel Implementation of Genetic Algorithms in a Classifier System, Proc. Int'l Conference on Genetic Algorithms and their Applications.

No context found.

G. Robertson. Parallel implementation of genetic algorithms in a classifier system. Genetic algorithms and simulated annealing, L.Davis ed., Morgan Kaufmann Pub., pages 129--140, 1987.

No context found.

Robertson, G. (1987). Parallel implementations of genetic algorithms in a classifier system. In J. J. Grefenstette, editor, Genetic Algorithms and Their Applications: Proceedings of the Second International Conference on Genetic Algorithms.

First 50 documents

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