A novel distributed genetic algorithm implementation with variable number of islands

Jumonji, Takuma; Chakraborty, Goutam; Mikami, Hiroshi; Matsuhara, Masafumi

doi:10.1109/cec.2007.4425088

Cited by 7 publications

(7 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [15], a Variable Island GA has been proposed where the initial number of islands has been set to one and each island has been populated by only two individuals. The decision of an island creation or removal has been taken based on the status of convergence and the fitness of their two individuals.…”

Section: Related Workmentioning

confidence: 99%

Scalable distributed evolutionary algorithm orchestration using Docker containers

Dziurzański

Zhao

Przewoźniczek

et al. 2020

Journal of Computational Science

View full text Add to dashboard Cite

In smart factories, integrated optimisation of manufacturing process planning and scheduling leads to better results than a traditional sequential approach but is computationally more expensive and thus difficult to be applied to real-world manufacturing scenarios. In this paper, a working approach for cloud-based distributed optimisation for process planning and scheduling is presented. Three managers dynamically governing the creation and deletion of subpopulations (islands) evolved by a multi-objective genetic algorithm are proposed, compared and contrasted. A number of test cases based on two real-world manufacturing scenarios are used to show the applicability of the proposed solution.

show abstract

Section: Related Workmentioning

confidence: 99%

Scalable distributed evolutionary algorithm orchestration using Docker containers

Dziurzański

Zhao

Przewoźniczek

et al. 2020

Journal of Computational Science

View full text Add to dashboard Cite

show abstract

“…The performance of SGA and EPDGA were compared based on the best fitness value and computation time. Rastrigin function is generally used to test and verify the efficiency of any minimizing algorithm [5] [7]. Following the Rastrigin function testing, the adaptive beam forming fitness function was tested to prove that the new algorithm performs better than standard GA in reducing the power usage at Node B.…”

Section: Simulationmentioning

confidence: 99%

“…It is known to be able to solve wide varieties of problems and the performance is better than other competitive methods [5]. But it has some potential drawbacks which become apparent while trying to solve complex multimodal problems.…”

Section: Introductionmentioning

confidence: 99%

“…But it has some potential drawbacks which become apparent while trying to solve complex multimodal problems. The main problems are the computation load of fitness calculation, the occurrence of early convergence and the difficulty to set parameter values effectively [5].…”

Section: Introductionmentioning

confidence: 99%

“…PDGA can avoid early convergence by maintaining diversity in population using migration. PDGA has better search performance and the computation time is shorter than SGA [5]. This model avoids the overhead of central control and presents a nature of parallelism, which makes it easy to implement this model on multiprocessor platforms and drastically speed up of evolution [9].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Embedded Parallel Distributed Artificial Intelligent Processors for Adaptive Beam Forming in WCDMA System

Krishnan

Kiong

Koh

et al. 2008

2008 6th National Conference on Telecommunication Technologies and 2008 2nd Malaysia Conference on Photonics

View full text Add to dashboard Cite

Genetic algorithms (GAs) are powerful search techniques that are used successfully to solve problems in many different disciplines. One application would be in WCDMA adaptive beam forming technique. Adaptive antenna has dynamic beam to cater for users' needs and provides better capacity for mobile communication but requires more intelligent and advance beam forming algorithm such as genetic algorithm. Compared to Standard GAs, Parallel Distributed GAs promise substantial gain in terms of convergence performance. In this paper, an embedded parallel and distributed genetic algorithm (EPDGA) with dynamic parameter setting on a multiprocessor system is proposed. The proposed scheme applies a master-slave architecture where the total active unit equipments (UE) are distributed to subpopulations (slaves) that evolve separately and exchange individuals occasionally. The power usage at Node B is used as fitness function to compare the performance of EPDGA and standard GA. Simulation results show that EPDGA converges faster and is better in adaptive antenna beam forming in the aspect of power usage at Node B as compared to standard GA.Keywords-Adaptive beam forming, parallel distributed genetic algorithm, master-slave architecture, WCDMA

show abstract

A Fast-Converging Particle Swarm Optimization through Targeted, Position-Mutated, Elitism (PSO-TPME)

Shaqarin

Noack

2023

Int J Comput Intell Syst

View full text Add to dashboard Cite

We improve convergence speed by two orders of magnitude and the global exploration capabilities of particle swarm optimization (PSO) through targeted position-mutated elitism (TPME). The proposed fast-converging TPME operator requires a fitness-based classification technique to categorize the particles. The introduced classification is motivated by its simplicity, low memory requirements, and automated termination criteria based on convergence. The three key innovations address particle classification, elitism, and mutation in the cognitive and social model. PSO-TPME is benchmarked against five popular PSO variants for multi-dimensional functions, which are extensively adopted in the optimization field, In particular, the convergence accuracy, convergence speed, and the capability to find global minima are investigated. The statistical error is assessed by numerous repetitions. The simulations confirmed that in ten of the thirteen investigated functions, the proposed PSO variant outperforms other variants in terms of convergence rate and accuracy by at least two orders of magnitude. On the other hand, the simulations demonstrated the early exploration capabilities of PSO-TPME in all tested functions. In the first ten iterations, PSO-TPME outperformed all the investigated PSO variants by at least two orders of magnitude.

show abstract

A novel distributed genetic algorithm implementation with variable number of islands

Cited by 7 publications

References 5 publications

Scalable distributed evolutionary algorithm orchestration using Docker containers

Scalable distributed evolutionary algorithm orchestration using Docker containers

Embedded Parallel Distributed Artificial Intelligent Processors for Adaptive Beam Forming in WCDMA System

A Fast-Converging Particle Swarm Optimization through Targeted, Position-Mutated, Elitism (PSO-TPME)

Contact Info

Product

Resources

About