Parallel Genetic Algorithm based on a new migration strategy

This paper proposed a new average non-dominated sorting genetic algorithm (NAVNSGA). This idea is inspired from the combination of non-elitist multi-objective evolutionary algorithms, elitist multiobjective evolutionary algorithms, and statistical calculations. The proposed NAVNSGA is improved the disadvantages of the Elitist multi-objective algorithms and Non-elitist multi-objective algorithms as possible. The NAVNSGA is compared with useful algorithms such as the non-elitist sorting genetic algorithm (NSGAI) and non-elitist sorting genetic algorithm (NSGAII) and the results obtained are showed the superiority of the proposed algorithm. Additionally, the NAVNSGA algorithm is combined with the concepts of the Game theory to propose a hybrid algorithm for determining Nash equilibrium in the game theory. The combination of the NAVNSGA algorithm with the game theory previously is used for improving engineering systems, such as I-beam designing. The results obtained are showed the advantage of the proposed algorithm with those reported in the literature.

show abstract

“…The equation (2) is a statistical method that uses for increasing diversity in [29]. If the condition in Eq.…”

Section: The Second Innovation In the Navnsga Algorithmmentioning

confidence: 99%

“…(2) is satisfied, some genes of chromosomes in the non-dominated set( the first class ) will be randomly reset. In fact, it is a Hill-Climbing method that is using for blind search [29]. The procedure of the NAVNSGA algorithm is as follows (Algorithm 5):…”

Section: The Second Innovation In the Navnsga Algorithmmentioning

confidence: 99%

A New Multi-objective Sorting Algorithm And Its Combination With Game Theory For Optimizing I-beam Engineering System

Navidi¹,

Khanesar²,

Falahiazar³

2015

J. Math. Computer Sci.

View full text Add to dashboard Cite

show abstract

“…Li and Huang [5] conducted research on parallel genetic algorithms, analyzed occurred shortcomings and solved them with an adaptive migration strategy technique by adjusting the number of migrants using individual function similarity. Falahiazar et al [6] made modifications on the parallel genetic algorithms for a better migration strategy. The technique used the max-min and a hill-climbing algorithm to structure the genetic algorithm migration.…”

Section: Introductionmentioning

confidence: 99%

A coarse-grained parallelization of genetic algorithms

Rathomi

Pulungan

2018

Int. J. Adv. Intell. Informatics

View full text Add to dashboard Cite

Genetic algorithms are frequently used to solve optimization problems. However, the problems become increasingly complex and time consuming. One solution to speed up the genetic algorithm processing is to use parallelization. The proposed parallelization method is coarse-grained and employs two levels of parallelization: message passing with MPI and Single Instruction Multiple Threads with GPU. Experimental results show that the accuracy of the proposed approach is similar to the sequential genetic algorithm. Parallelization with coarse-grained method, however, can improve the processing and convergence speed of genetic algorithms.

show abstract

“…Since the brain has various performances, learning process can be complex and different as well. Neural networks trained with gradient descent based algorithms [5] and evolutionary algorithms [6][7]. In this regard, not only the considerable importance of learning fields and material, but also learning methodologies are being stressed.…”

Section: Introductionmentioning

confidence: 99%

An Improved Model Of Brain Emotional Learning Algorithm Based On Interval Knowledge

Sharafi

Setayeshi

Falahiazar

2015

J. Math. Computer Sci.

View full text Add to dashboard Cite

The brain emotional learning algorithm inspired by a reduced system of a computational model simulates the brain learning performance quite simply with mimicking mammalian brains. The present paper endeavors to come forward to an improved model of the emotional learning algorithm based on the interval knowledge. In this proposed model, based on the interval knowledge, the weights of the amygdala and orbitofrontal sections will be updated. Eventually, the results of implementing and performing the improved brain emotional learning algorithm will be presented to be compared with the original version of the algorithm to Prediction the chaotic time series, Lorenz and Rossler, about which a noticeable improvement in its precision, accuracy and speed of convergence of the final results is reported.

show abstract

Parallel Genetic Algorithm based on a new migration strategy

Cited by 8 publications

References 10 publications

A New Multi-objective Sorting Algorithm And Its Combination With Game Theory For Optimizing I-beam Engineering System

A New Multi-objective Sorting Algorithm And Its Combination With Game Theory For Optimizing I-beam Engineering System

A coarse-grained parallelization of genetic algorithms

An Improved Model Of Brain Emotional Learning Algorithm Based On Interval Knowledge

Contact Info

Product

Resources

About