Cohoon, J., Hegde, S., Martin, W., and Richards, D. (1987). Punctuated equilibria: a parallel genetic algorithm. In Grefenstette, J., editor, Proceedings of the Second International Conference on Genetic Algorithms, pages 148--154, Hillsdale, NJ, USA. Lawrence Erlbaum Associates.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....sequential search programs. The primary rationale behind sharing information among search programs is to speed up the exploration of the solution space. However, several researchers have observed that cooperation by sharing information was more than just hardware acceleration (Cohoon all [5]) Researchers in the eld of parallel genetic algorithms (Tanase [35,36] Jog and Gucht [21] Cohoon all [5] Munetomo, Takai Sato [31] concluded that cooperation not only produces speed up, but it also profoundly modi es the search pattern of the cooperating programs. They found empirical ....

....speed up the exploration of the solution space. However, several researchers have observed that cooperation by sharing information was more than just hardware acceleration (Cohoon all [5] Researchers in the eld of parallel genetic algorithms (Tanase [35,36] Jog and Gucht [21] Cohoon all [5], Munetomo, Takai Sato [31] concluded that cooperation not only produces speed up, but it also profoundly modi es the search pattern of the cooperating programs. They found empirical evidence that the speci cation of how search processes cooperate has an impact on how PGA succeed in nding ....

J. Cohoon, S. Hedge, W. Martin, and D. Richards. Punctuated Equilibria: A Parallel Genetic Algorithm. In J.J. Grefenstette, editor, Proceedings of the Second International Conference on Genetic Algorithms and their Applications, pages 148-154. Lawrence Erlbaum Associates Publishers, 1987.

....carried out in parallel using discrete time steps. Selection and crossover occurs locally within small, overlapping neighbourhoods. Subsequently, good genes slowly diffuse across the lattice. The benefits of isolating subpopulation have long been known in the field of evolutionary algorithms [7] [9]. Parallel GAs are known to handle difficult multimodal functions more efficiently than serial GAs [10] However, both the course and fine grained models are artificially constrained. The quality of the search and efficiency of the algorithms can be severely affected by the parallel parameter ....

Cohoon, J.P., Hegde, S.U., Martin,W.N. and Richards, D. (1987), Punctuated Equilibria: a Parallel Genetic Algorithm. In J. Grefenstette (ed.). Proceedings of the Second International Conference on Genetic Algorithm, Lawrence Erlbaum Associates, pp. 148-154.

....migration was used than in the completely isolated cases. Surprisingly, convergence was faster when a high migration rate was used (r = 0:5) than with a low rate. Also in 1987, Cohoon, Hedge, Martin, and Richards proposed an implementation of a PGA based on the theory of punctuated equilibria [18]. One aspect of the punctuated equilibria theory is that new species are likely to form quickly in relatively small isolated populations after some change in the environment occurs. The genetic composition of a population is an important part of an organism environment and thus, immigration can be ....

.... was introduced in a 1990 paper presented at the first Parallel Problem Solving from Nature workshop (PPSN) where migration occurs after the subpopulations converge (the author uses the term degenerate ) 17] The same concept was later used (with some alterations) by Munetomo, Takai, and Sato [18]. These researchers raised an important (and yet unanswered) question: when is the right time to migrate If migration occurs too early the number of correct building blocks in the migrants may be too low to influence the search on the right direction and expensive communication resources would be ....

J. Cohoon, S. Hegde, W. Martin, and D. Richards. Punctuated equilibria: a parallel genetic algorithm. In J.J. Grefenstette, editor, Proceedings of the Second International Conference on Genetic Algorithms. Lawrence Erlbaum Associates, 1987.

....exist to parallelize a genetic algorithm [2] However, most of these methods result only in a speed up of the algorithm without qualitative improvements to the problem solutions. To gain better problem solutions, we use the theory of punctuated equilibria to design a parallel genetic algorithm [7], 10] A genetic algorithm with punctuated equilibria is a parallel genetic algorithm in which independent subpopulations of individuals with their own fitness functions evolve in isolation except for an exchange of individuals (migration) when a state of equilibrium throughout all the ....

.... for an exchange of individuals (migration) when a state of equilibrium throughout all the subpopulations has been reached (see Figure 3) Previous research has shown genetic algorithms with such punctuated equilibria to have superior performance when compared to sequential genetic approaches [7], 9] The parallel structure of our algorithm for the case of nine processors is shown in Figure 4. We assign a set of n individuals (problem solutions) to each of the N processors, for a total population size of n Theta N . The set assigned to each processor, c, is its subpopulation, P c . The ....

J. P. Cohoon, S. U. Hedge, W. N. Martin, and D. S. Richards, "Punctuated Equilibria: A Parallel Genetic Algorithm," Proc. Second International Conference on Genetic Algorithms, pp. 148-154, 1987.

....exist to parallelize a genetic algorithm [2] However, most of these methods result only in a speed up of the algorithm without qualitative improvements to the problem solutions. To gain better problem solutions, we use the theory of punctuated equilibria to design a parallel genetic algorithm [5], 9] A genetic algorithm with punctuated equilibria is a parallel genetic algorithm in which independent subpopulations of individuals with their own fitness function evolve in isolation, except for an exchange of individuals (migration) when a state of equilibrium throughout all the ....

.... except for an exchange of individuals (migration) when a state of equilibrium throughout all the subpopulations has been reached (see Figure 1) Previous research has shown genetic algorithms with punctuated equilibria to have superior performance when compared to sequential genetic approaches [5], 7] Migration Result Figure 1: Punctuated equilibria model. The parallel structure of our algorithm for the case of nine processors is shown in Figure 2. We assign a set of n individuals (problem solutions) to each of the N processors, for a total population size of n Theta N . The set ....

J.P. Cohoon, S.U. Hedge, W.N. Martin and D.S. Richards, "Punctuated Equilibria: A Parallel Genetic Algorithm", Proc. Second International Conf. on Genetic Algorithms, pp. 148154, 1987.

....are sequential approaches, i.e. one population evolves by means of genetic operators. However, recent publications indicate that parallel genetic algorithms with isolated evolving subpopulations (that exchange individuals from time to time) may offer advantages over sequential approaches [2] [7], 9] 20] 23] We present a parallel genetic algorithm for detailed routing, called GAP (Genetic Algorithm with Punctuated equilibria) that runs on a distributed network of workstations. To our knowledge, this is the first approach which includes crosstalk considerations directly in a gridded ....

....a genetic algorithm [2] However, most of these methods result only in a speed up of the algorithm without qualitative improvements to the problem solutions. To gain better problem solutions, we designed a parallel genetic algorithm inspired by concepts from the theory of punctuated equilibria [7], 12] A genetic algorithm with punctuated equilibria is a parallel genetic algorithm in which independent subpopulations of individuals with their own fitness functions evolve in isolation, except for an exchange of individuals (migration) when a state of equilibrium throughout all the ....

[Article contains additional citation context not shown here]

J. P. Cohoon, S. U. Hedge, W. N. Martin, and D. S. Richards, "Punctuated Equilibria: A Parallel Genetic Algorithm," in Proc. Second Int. Conf. on Genetic Algorithms, 1987,

....of migrants leads to the behavior of the island model similar to the behavior of a panmitic model. A few migrants prevent the GA from mixing the genotypes, and thus reduce the possibility to bypass the local optima inside the islands. Implementations of coarse grained parallel GAs can be found in [18, 19,20,21,22,23]. To implement Cga LK the island model was adopted. Moreover, Cga LK exploits the MPI message passing library [24] and the SPMD programming model [25] According to this programming model all the processing nodes run the same code which simulates a di#erent population. We can consider each ....

S. Cohoon, J. Hedge, S. Martin, and D. Richards. Punctuated equilibria: a parallel genetic algorithm. IEEE Transaction on CAD, 10(4):483--491, April 1991.

....leads to the behavior of the island model similar to the behavior of a panmitic model. A few migrants prevent the GA from mixing the genotypes, and thus reduce the possibility to bypass the local optimum value inside the islands. Implementations of coarse grained parallel GAs can be found in [10, 20, 21, 4, 18, 16]. 3 Designing parallel GAs We implemented both fine Gammagrained and coarse Gammagrained parallel GAs applied to the classic Traveling Salesman Problem on a 128 node nCUBE 2 hypercube. Their performance was measured by varying the type and value of some genetic operators. In the following ....

....phase, and the migration of the best individuals among sub population is performed among ring connected islands. Each of the P processors manages N=P individuals, with N population size (640 individuals in our case) The number of migrants is a fixed percentage of the sub population. As in [4], migration occurs periodically in a regular time rhythm, after a fixed number of generations. In order to include all the migrants in the current sub populations, and to merge the subpopulation with the locally generated solutions, R1 and R2 replacement criteria were used, respectively. ....

S. Cohoon, J. Hedge, S. Martin, and D. Richards. Punctuated equilibria: a parallel genetic algorithm. IEEE Transaction on CAD, 10(4):483--491, April 1991.

....318 and 532 cities problems. A typical result is shown in Figure 12. 7 Parallel GA towards Memetic Algorithms Due to its intrinsic parallelism and the fact that multiprocessor architectures are each day more available, there is considerable interest in the GA community to exploit this advantage [55] [56] 99] 150] 151] 186] 187] I will not discuss these in particular. Instead, I will concentrate in some parallel GA that have turned into examples of memetic algorithms. SAGA A Parallel Genetic Heuristic for the Quadratic Assignment Problem SAGA is one of the two parallel heuristics ....

J.P. Cohoon et al., "Punctuated Equilibria: A Parallel Genetic Algorithm ", Genetic Algorithms and their Applications, J.J. Grefenstette, editor, Lawrence Erlbaum Associated (1987).

....use other criteria for selection and replacement. 5 Migration selection. In most cases the best or a limited sequence of the best individuals are selected for migration. It is also possible to choose a fitness proportional or rankproportional selection scheme as it is used for mating selection. COHOON et al. 1987) follow a different approach: There in each migration phase several individuals are selected and replaced randomly. Migration replacement. In most cases only the worst or a limited sequence of the worst individuals are replaced. Similar to the migration selection, a fitness proportional or ....

COHOON, J. P., HEGDE, S. U., MARTIN, W. N., & RICHARDS, D. S. 1987. Punctuated Equilibria: A Parallel Genetic Algorithm. In: (GREFENSTETTE, 1987), 148--154.

....leads to the behavior of the island model similar to the behavior of a panmitic model. A few migrants prevent the GA from mixing the genotypes, and thus reduce the possibility to bypass the local optimum value inside the islands. Implementations of coarse grained parallel GAs can be found in [10, 20, 21, 4, 18, 16]. 3 Designing parallel GAs We implemented both fine Gamma grained and coarse Gamma grained parallel GAs applied to the classic Traveling Salesman Problem on a 128 node nCUBE 2 hypercube. Their performance was measured by varying the type and value of some genetic operators. In the following ....

....phase, and the migration of the best individuals among sub population is performed among ring connected islands. Each of the P processors manages N=P individuals, with N population size (640 individuals in our case) The number of migrants is a fixed percentage of the sub population. As in [4], migration occurs periodically in a regular time rhythm, after a fixed number of generations. In order to include all the migrants in the current sub populations, and to merge the sub population with the locally generated solutions, R1 and R2 replacement criteria were used, respectively. ....

S. Cohoon, J. Hedge, S. Martin, and D. Richards. Punctuated equilibria: a parallel genetic algorithm. IEEE Transaction on CAD, 10(4):483--491, April 1991.

....island model or migration model is used for an EA with panmictic subpopulations. The neighborhood model is sometimes also called diffusion model. In the past, work on population structures always emphasized a particular model. For panmictic subpopulations, there are mostly empirical studies, e.g. [2, 3, 4, 5]. In a theoretical approach, Cant u Paz [6] presented optimal subpopulation sizes for some special instances of Genetic Algorithms. 2 Neighborhood models were analyzed with respect to local selection schemes [7, 8] as well as neighborhood shapes in grid topologies [9, 10] 3 A Hypergraph Based ....

J. P. Cohoon, S. U. Hedge, W. N. Martin, and D. S. Richards. Punctuated Equilibria: A Parallel Genetic Algorithm. In Grefenstette [20], pages 148--154.

....[308, 309] Chiang, Hsaio Dong, 224, 245, 281] Chiang, Hsiao Dong, 134] Chipper eld, Andrew J. 131] Chockalingam, T. 74] Chopard, Bastien, 115, 186, 531] Christou, I. T. 238] Chung, M. J. 216] Cierniak, M. 158] Ciesla, W. 233, 311] Cline, D. D. 474, 475] Cohoon, James P. [374, 375, 376, 377, 378, 379] Collins, Robert James, 476] Colombetti, Marco, 15, 389, 391] Compiani, Mario, 199] Conrad, Michael, 34] Corno, Fulvio, 239, 280] Authors 15 Corwin, Edward M. 396] Crummey, T. P. 131] Crutch eld, James P. 12, 16, 20, 23, 41, 45, 46] Cui, Jun, 547, 399, 548, 400] Daemi, M. F. ....

....41] Harris, F. C. 204] Harrison, R. F. 42] Hart, William E. 300] Hart, William Eugene, 95] H artfelder, Michael, 402] Hartley, Stephen J. 372] Harvey, N. R. 175] Hauser, R. 96, 170] He, Q. H. 345] 16 Distributed genetic algorithms Hedetniemi, S. M. 310] Hegde, Shailesh U. [374, 375, 376, 379, 414] Henry, Kevin, 11] Heraj arvi, Juha, 57] Herrmann, Frank, 342] Higuchi, T. 415] Hines, J. 204] Hirata, H. 21] Hirayama, K. 248] Hirose, Tetsuya, 427] Ho, Lap Wah Lawrence, 321] Ho meister, Frank, 416, 417, 487] Hofmann, Hartmut M. 336, 337] Horrocks, David H. 271] Hosticka, ....

[Article contains additional citation context not shown here]

James P. Cohoon, Shailesh U. Hegde, Worthy N. Martin, and Dana S. Richards. Punctuated equilibria: a parallel genetic algorithm. In Grefenstette [565], pages 148-154. ga:Cohoon87b.

.... to which a genome migrates is chosen stochastically and asynchronously (e.g. Norman, 1988) in others deterministically in rotation (e.g. Whitley et al. 1989) while in still others the islands themselves have a structure such as a ring and migrations only occur between neighbouring islands (e.g. Cohoon et al. 1987); this last case is sometimes known as the stepping stone model. The largely independent course of evolution on each island again encourages niching (or speciation) while ultimately allowing genetic information to migrate anywhere in the (structured) population, and again this helps to avoid ....

Cohoon et al., 1987. J. P. Cohoon, S. U. Hegde, W. N. Martin, and D. Richards. Punctuated equilibria: a parallel genetic algorithm. In Proceedings of the Second International Conference on Genetic Algorithms. Lawrence Erlbaum Associates (Hillsdale, New Jersey), 1987.

....one might consider an island model in which each island is structured as a grid of locally interacting individuals. For a recent review of parallel evolutionary algorithms (including several references) the reader is referred to [16] Though such parallel models have empirically proven worthwhile [1, 4, 7, 8, 10, 15, 17], there seems to be lacking a better understanding of their workings. Gaining insight into the mechanisms of parallel evolutionary algorithms calls for the introduction of statistical measures of analysis. This is the underlying motivation of our paper. Specifically, concentrating on cellular ....

J. P. Cohoon, S. U. Hedge, W. N. Martin, and D. Richards. Punctuated equilibria: A parallel genetic algorithm. In J. J. Grefenstette, editor, Proceedings of the Second International Conference on Genetic Algorithms, page 148. Lawrence Erlbaum Associates, 1987.

....then it is likely that the subpopulations will converge on different species. The exchange step might be seen as species migration caused by a catastrophic environmental shift. Such an interpretation of Grefenstette s network model, called punctuated equilibria, is described by Cohoon et al. [22] . A fine grained variant of GALO s network implementation is Muhlenbein s PGA [23] which maintains a single structure on each processor and uses breeding for information exchange. Each of the models has its advantages, depending on the circumstances. If the execution times for the local ....

J. Cohoon, S. Hegde, W. Martin, and D. Richards, 1987. Punctuated Equilibria: a Parallel Genetic Algorithm, In Genetic Algorithms and Their Applications: Proceedings of the Second International Conference on Genetic Algorithms, 177183.

.... to which a genome migrates is chosen stochastically and asynchronously (e.g. Norman, 1988) in others deterministically in rotation (e.g. Whitley et al. 1989) while in still others the islands themselves have a structure such as a ring and migrations only occur between neighbouring islands (e.g. Cohoon et al. 1987); this last case is sometimes known as the stepping stone model. The largely independent course of evolution on each island again encourages niching (or speciation) while ultimately allowing genetic information to migrate anywhere in the (structured) population, and again this helps to avoid ....

J. P. Cohoon, S. U. Hegde, W. N. Martin, and D. Richards, 1987. Punctuated equilibria: a parallel genetic algorithm. In Proceedings of the Second International Conference on Genetic Algorithms. Lawrence Erlbaum Associates (Hillsdale, New Jersey).

....to the spatial distribution of population members [21] 2] At one end of the spectrum are so called coarse grained parallel GAs. In this case, the global population is divided into a few distinct non overlapping subpopulations. Communication among them is accomplished by periodic migration [3]. Selection within a subpopulation is an issue but there are no inter deme selection issues. If a parallel computer is applied to coarse grained GA simulation, a MIMD architecture is most appropriate [22] Rigorous experimentation with coarse grained GAs was first performed by Tanese [23] She ....

Cohoon, J. P., Hegde, S. U., Martin, W. N., and Richards, D. "Punctuated equilibria: a parallel genetic algorithm", in [8], pp. 148--154.

....As with SA, the GA exhibits elements of both randomization and local search. Likewise, it has proven to be quite effective for approximating global solutions to many types of NP hard combinatorial problems, and indeed has been successfully applied to a number of VLSI design automation problems [15, 16, 17, 23, 33, 64, 81]. Section 4.2.1 introduces the terminology of genetic algorithms as inspired by the analogy with natural evolution. Section 4.2.2 introduces the original simple genetic algorithm of Holland [45] Section 4.2.3 provides a theoretical basis for the operation of simple genetic algorithm with the ....

....the available processors with minimum communications overhead is needed. The population distribution method commonly used by researchers utilizing GAs on distributed architectures is typically some subspeciation technique, in which each processor independently evolves a subpopulation of strings [8, 15, 16, 17, 23, 33]. Since larger populations tend to produce better GA solution quality through improved diversity, all of these population distribution schemes involve some mechanism for the migration of solutions between processors. This allows for a rediversification of solutions in the subpopulations which can ....

J.P. Cohoon, S.U. Hegde, W.N. Martin, and D.S. Richards, "Punctuated Equilibria: A Parallel Genetic Algorithm," Proc. Second Int. Conf. Genetic Algs.and Their Apps., Cambridge, MA, 148-154, 1987.

No context found.

Cohoon, J., Hegde, S., Martin, W., and Richards, D. (1987). Punctuated equilibria: a parallel genetic algorithm. In Grefenstette, J., editor, Proceedings of the Second International Conference on Genetic Algorithms, pages 148--154, Hillsdale, NJ, USA. Lawrence Erlbaum Associates.

No context found.

J. Cohoon, S. Hedge, W. Martin, and D. Richards. Punctuated Equilibria: A Parallel Genetic Algorithm. In J.J. Grefenstette, editor, Proceeding of the Second International Conference on Genetic Algorithms and their Applications, pages 148--154. Lawrence Erlbaum Associates, Hillsdale, NJ, 1987. 37

No context found.

Cohoon, J. P., Hegde, S. U., Martin, W. N., and Richards, D. "Punctuated equilibria: A parallel genetic algorithm". In: Proceedings of the Second International Conference on Genetic Algorithms. Lawrence Erlbaum Associates (1987) 148-154.

No context found.

J.P. Cohoon, S.U. Hegde, W.N. Martin and D. Richards, Punctuated Equilibria: A Parallel Genetic Algorithm, Proceedings of the Second International Conference on Genetic Algorithms and their Applications, J.J. Grefenstette (editor), July 1987, pp. 148--154. PARALLEL METAHEURISTICS FOR COMBINATORIAL OPTIMIZATION 17

No context found.

J. P. Cohoon, S. U. Hedge, W. N. Martin, and D. Richards, Punctuated equilibria: a parallel genetic algorithm, in Proceedings of the Second International Conference on Genetic Algorithms, J. J. Grefenstette ed., Lawrence Erlbaum Associates, 1987, pp.148--154.

No context found.

J.P. Cohoon et al., Punctuated Equilibria: A Parallel Genetic Algorithm, in: J.J. Grefenstette, ed., Proc. Second Intl. Conf. on Genetic Algorithms and their Applications, (Lawrence Erlbaum Associates, Hillsdale, New Jersey,1987) 148-154.

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