GORGES-SCHLEUTER, M. 1992. Comparison of Local Mating Strategies in Massively Parallel Genetic Algorithms. In: (M ANNER & MANDERICK, 1992), 553--562.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....between the above classification scheme and earlier classifications. 2 Reference Global Model Regional Model Local Model GREFENSTETTE (1981) Master Slave Network MANDERICK et al. 1989) R Algorithm Coarse Grain Fine Grain MACFARLANE et al. 1990) Farming Migration Diffusion GORGES SCHLEUTER (1992) Panmixia Migration Model Diffusion Model DORIGO et al. 1993) Island Model Neighbourhood M. WHITLEY (1993) Global Pop. Island Model Cellular GA CANT U PAZ (1995) Global Par. Coarse Grain Fine Grain Figure 1: Summary of classification schemes for parallel population models 3 Global ....

GORGES-SCHLEUTER, M. 1992. Comparison of Local Mating Strategies in Massively Parallel Genetic Algorithms. In: (M ANNER & MANDERICK, 1992), 553--562.

....local selection within subpopulations . For two of these algorithmic variants, migration models [64] and diffusion models [30,76] it has been observed that local selection techniques not only yield a considerable speed up on parallel computers, but also improve the robustness of the algorithms [63,31] and ease the search for the whole Pareto optimal subset x ae S in case of multiple criteria [46] Since a detailed presentation of parallel EAs are beyond the scope of this article, the reader is referred to [38] 13 3 NUCLEAR REACTOR CORE RELOAD DESIGN 3.1 Problem Description The following ....

M. Gorges-Schleuter. Comparison of local mating strategies in massively parallel genetic algorithms. In Manner and Manderick [49], pages 553--562.

....all the individuals and selection, reproduction and mating take place locally within the neighbourhood. Information slowly diffuses across the grid thus clusters of solutions are formed around different optima. Recent researches proved that such an approach shows no degrade of the solution quality (Gorges Schleuter 1992) and an improved performance. Cellular automata (CAs) Toffoli al. 1986) can be used as a framework to enable a fine grained parallel implementation of GP through the diffusion model. A CA is composed of a set of cells in a regular spatial lattice, either one dimensional or multidimensional. ....

M. Gorges-Schleuter (1992). Comparison of local mating strategies in massively parallel genetic algorithms, in R. Manner and B. Manderick (eds.) Parallel Problem solving from Nature II. North Holland, pp. 554561.

....212, 282] Gitler, Daniel, 211] Giusti, Giuliano, 86, 199] Godart, C. 176] Godza, G. 283] Gogonea, Valentin, 236] Gold, S onke Sonnich, 488] Goldberg, David E. 210, 312] Goodman, Erik D. 88, 188, 189, 242, 341, 483] Gordon, Vahl Scott, 135, 148, 520] Gorges Schleuter, Martina, [93, 405, 406, 407, 408, 409, 410, 404, 442, 76] Grabski, W. 122] Green, David G. 22] Green, P. R. 116, 145] Greenwood, Garrison W. 136] Grefenstette, John J. 468] Grocholewska Czurylo, A. 28] Gruau, Fr ed eric C. 411] Gucht, Dirk Van, 58] Guidec, Fr ed eric, 237, 275, 72] Gupta, Ajay, 136] Hajela, Prabhat, 206, 286] ....

.... [109] k partition problem, 378] knapsack problem, 515] laminates, 189] LAN, 485, 136] lattice model 27 mer, 188] 64 mer, 188] layout design, 375, 379, 418, 523, 438, 246, 262, 330] VLSI, 294] learning, 546] LGANN, 199] LibGA, 127] load balancing, 484, 485, 472] local mating, [409] machine learning, 483, 484, 485, 452, 92, 159, 212, 34] machine sequencing, 349] macro cell layout, 488] majority classi cation, 26] manufacturing, 257, 344] cells, 293] mapping, 122] mapping problem, 461, 502] medicine classi cation, 430] messy GA, 440, 439, 91, 111] meta GA, 402] ....

[Article contains additional citation context not shown here]

Martina Gorges-Schleuter. Comparison of local mating strategies in massively parallel genetic algorithms. In Manner and Manderick [563], pages 553-562. y ga:Gorges-Schleuter92b.

....Glasmacher, Klaus, 238] Glorennec, Pierre Yves, 243] Glower, J. S. 244] Goldberg, David E. 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 3, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265] Gonzales Seco, Jose, 344] Gordon, Vahl Scott, 693] Gorges Schleuter, Martina, [266, 267] Gottvald, A. 268] Graudenz, Dirk, 108] Greene, David Perry, 269] Grefenstette, John J. 156, 157, 270, 271, 272, 273] Grierson, D. E. 541] Grosmann, Tal, 143] Gruau, Fr ed eric C. 275, 276] Guan, Shanguchuan, 326] Guo, Zhichao, 277, 278] Gupta, Akhil Datta, 454] Gurdal, ....

.... laminates, 116, 285, 286] lasers, 363] layout design, 309, 319, 320, 340, 472, 474, 550, 661, 662] learning, 66, 71, 167, 242, 452, 547, 611, 636, 677, 678, 717] learning optimisation, 622, 623] line balancing, 187] load balancing, 551] local hill climbing, 707] local mating, [267] local search, 48] logic, 269] long path problems, 257] machine learning, 47, 109, 127, 152, 156, 168, 170, 172, 270, 331, 348, 349, 428, 456, 563, 606, 668, 671, 708, 709] machine learning Subject index 23 decision making, 383] review, 632] manipulator design, 381] manufacturing ....

[Article contains additional citation context not shown here]

Martina Gorges-Schleuter. Comparison of local mating strategies in massively parallel genetic algorithms. In Manner and Manderick [539], pages 553--562. y ga:Gorges-Schleuter92b.

....fine grained and coarse grained approaches. Characteristic of fine grained approaches is that very few, often only one, individuals are assigned to one processors and individuals are connected by a population structure. An example of an implementation of such an approach for the TSP is ASPARAGOS [11] that was able to find very good solutions to the TSP instance att532. A typical example of a coarse grained approach is the island genetic algorithm in which the population of a genetic algorithm is divided into several subpopulations. Communication takes place by exchanging individuals at ....

M. Gorges-Schleuter. Comparison of Local Mating Strategies in Massively Parallel Genetic Algorithms. In PPSN-II, pages 553--562, 1992.

....used a population structure that resembles a ladder with the upper and lower ends tied together. ASPARAGOS was used to solve some difficult combinatorial optimization problems with great success. Later, a linear population structure was used [Gor 91] and different mating strategies were compared [Gor 92b] ASPARAGOS uses a local hillclimbing algorithm to improve the fitness of the individuals in its population. This makes it difficult to isolate the contribution of the spatial structure of the population to the search quality. However, all the empirical results show that ASPARAGOS is is a very ....

GORGES-SCHLEUTER M., Comparison of local mating strategies in massively parallel genetic algorithms ». In MÄNNER R., MANDERICK B., Eds., Parallel Problem Solving from Nature, 2, p. 553--562, Elsevier Science (Amsterdam), 1992.

....the diffusion model fits the restrictions of massively parallel environments. A combined approach of traditional GA s and diffusion models is shown by the ECO framework of Davidor [4] A brief outline of other local mating strategies proposed is published in a recent paper of Gorges Schleuter [8]. 3 Incorporating Patterns of Social Behavior In structured populations derived from diffusion models mating is restricted to a small number of nearby individuals. Hence global premature convergence is alleviated at the expense of incest in the neighborhood. While the environment changes ....

Gorges-Schleuter, M.: Comparison of Local Mating Strategies in Massively Parallel Genetic Algorithms. Proc. of PPSN 2, North-Holland (1992) 553--562

....structure that looks like a ladder with the upper and lower ends tied together. ASPARAGOS was used to solve some difficult combinatorial optimisation problems with great success. Later, a linear population structure was tried (Gorges Schleuter 1991) and different mating strategies were compared (Gorges Schleuter 1992). ASPARAGOS uses a local hillclimbing algorithm to refine the individuals in its population. It is difficult to isolate the contribution of the spatial structure of the population on the quality of the search from the contribution form the hillclimber. However, all the empirical results gathered ....

Gorges-Schleuter, M. (1992). Comparison of local mating strategies in massively parallel genetic algorithms, in R. Manner and B. Manderick (eds), Parallel Problem Solving from Nature, 2, Amsterdam: Elsevier Science, pp. 553--562.

....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 Internet. This approach allows anyone with access to the Internet to harness the huge number of unused CPU cycles the Internet contains to ....

Gorges-Schleuter, M. (1992). Comparison of local mating strategies in massively parallel genetic algorithms, in R. Manner and B. Manderick (eds), Parallel Problem Solving from Nature, 2, Amsterdam: Elsevier Science, pp. 553--562.

....acceptance function a( already described in (2) can be chosen. It remains to clarify how to choose a neighbor when recombination is desired. In this implementation two variants have been tried: Choose a neighbor at random or choose the best one of the neighborhood. Other variants can be imagined (Gorges Schleuter 1992). Similarly, many recombination operators are possible. Here, the following variant (hypercube recombination) has been used: Let x; y 2 IR n be the parent and v 2 IR n be the result of recombination. Then v = x (y Gamma x) Delta , where is a random vector uniformly distributed in [0; ....

Gorges-Schleuter, M. (1992). Comparison of local mating strategies in massively parallel genetic algorithms. In R. Manner and B. Manderick (Eds.), Parallel Problem Solving from Nature, 2, Amsterdam: North Holland, pp. 553--562.

....fine grained and coarse grained approaches. Characteristic of fine grained approaches is that very few, often only one, individuals are assigned to one processors and individuals are connected by a population structure. An example of an implementation of such an approach for the TSP is ASPARAGOS [12] that was able to find very good solutions to the TSP instance att532. A typical example of a coarse grained approach is the island genetic algorithm in which the population of a genetic algorithm is divided into several subpopulations. Communication takes place by exchanging individuals at ....

M. Gorges-Schleuter. Comparison of Local Mating Strategies in Massively Parallel Genetic Algorithms. In PPSN-II, pages 553--562, 1992.

....not able to observe significant permanent local differentiation with reasonable population sizes. Theoretical investigations in terms of the inbreeding coefficient performed gave evidence that the placement of the individuals along a ring might be an even better choice for the population structure [9]. The advantage is two fold: firstly, we find even higher local differentiation compared to the ladder population structure thus preventing the population to converge early to a single solution and secondly, we can reduce the population size without loss of quality and thus have the gain of faster ....

M. Gorges-Schleuter, Comparison of Local Mating Strategies in Massively Parallel GAs. In Proc 2nd PPSN, Brussels, 1992

....permanent local differentiation with reasonable population sizes. Theoretical investigations in terms of the inbreeding coefficient performed gave evidence that the placement of the individuals along a linear structure (ring, ladder) might be a better choice for the population structure [9]. The advantage is two fold: firstly, we find higher local differentiation compared to the torus structure thus preventing the population to converge early to a single solution and secondly, we can reduce the population size without loss of quality and thus have the gain of faster development. for ....

M. Gorges-Schleuter, Comparison of Local Mating Strategies in Massively Parallel GAs. In Proc PPSN 2, Elsevier, 1992, 553--562

No context found.

Gorges-Schleuter, M. (1992). Comparison of local mating strategies in massively parallel genetic algorithms. In R. M¨anner and B. Manderick, editors, Parallel Problem Solving from Nature II, pages 553--562, Amsterdam: North-Holland.

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