The Effect of Oscillating Population Size and Re-initialization on the Performance of Genetic Algorithms

2009 ◽  
Author(s):  
V.K. Koumousis ◽  
C.P. Katsaras
Keyword(s):  
Genetic Algorithms ◽  
Population Size

scholarly journals Training Multilayer Perceptron with Genetic Algorithms and Particle Swarm Optimization for Modeling Stock Price Index Prediction

Entropy ◽  
2020 ◽  
Vol 22 (11) ◽  
pp. 1239
Author(s):  
Fatih Ecer ◽  
Sina Ardabili ◽  
Shahab S. Band ◽  
Amir Mosavi
Keyword(s):  
Genetic Algorithms ◽  
Particle Swarm Optimization ◽  
Population Size ◽  
Correlation Coefficient ◽  
Multilayer Perceptron ◽  
Gaussian Function ◽  
Particle Swarm ◽  
Series Data ◽  
Output Function ◽  
Swarm Optimization

Predicting stock market (SM) trends is an issue of great interest among researchers, investors and traders since the successful prediction of SMs’ direction may promise various benefits. Because of the fairly nonlinear nature of the historical data, accurate estimation of the SM direction is a rather challenging issue. The aim of this study is to present a novel machine learning (ML) model to forecast the movement of the Borsa Istanbul (BIST) 100 index. Modeling was performed by multilayer perceptron–genetic algorithms (MLP–GA) and multilayer perceptron–particle swarm optimization (MLP–PSO) in two scenarios considering Tanh (x) and the default Gaussian function as the output function. The historical financial time series data utilized in this research is from 1996 to 2020, consisting of nine technical indicators. Results are assessed using Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE) and correlation coefficient values to compare the accuracy and performance of the developed models. Based on the results, the involvement of the Tanh (x) as the output function, improved the accuracy of models compared with the default Gaussian function, significantly. MLP–PSO with population size 125, followed by MLP–GA with population size 50, provided higher accuracy for testing, reporting RMSE of 0.732583 and 0.733063, MAPE of 28.16%, 29.09% and correlation coefficient of 0.694 and 0.695, respectively. According to the results, using the hybrid ML method could successfully improve the prediction accuracy.

Genetic algorithms with shrinking population size

2010 ◽  
Vol 25 (4) ◽  
pp. 691-705 ◽  
Author(s):  
Joshua W. Hallam ◽  
Olcay Akman ◽  
Füsun Akman
Keyword(s):  
Genetic Algorithms ◽  
Population Size

POPULATION SIZE MODELING FOR GA IN TIME-CRITICAL TASK SCHEDULING

2011 ◽  
Vol 22 (03) ◽  
pp. 603-620 ◽  
Author(s):  
WEI SUN
Keyword(s):  
Genetic Algorithms ◽  
Population Size ◽  
Task Scheduling ◽  
Scheduling Problems ◽  
Success Ratio ◽  
Critical Task ◽  
Time Critical

Genetic algorithms (GAs) have been well applied in solving scheduling problems and their performance advantages have also been recognized. However, practitioners are often troubled by parameters setting when they are tuning GAs. Population Size (PS) has been shown to greatly affect the efficiency of GAs. Although some population sizing models exist in the literature, reasonable population sizing for task scheduling is rarely observed. In this paper, based on the PS deciding model proposed by Harik, we present a model to represent the relation between the success ratio and the PS for the GA applied in time-critical task scheduling, in which the efficiency of GAs is more necessitated than in solving other kinds of problems. Our model only needs some parameters easy to know through proper simplifications and approximations. Hence, our model is applicable. Finally, our model is verified through experiments.

An Empirical Study of the Effect of Parameter Combination on the Performance of Genetic Algorithms

2013 ◽  
Vol 1 (2) ◽  
pp. 43-55
Author(s):  
Pi-Sheng Deng
Keyword(s):  
Genetic Algorithms ◽  
Empirical Study ◽  
Population Size ◽  
Mutation Rate ◽  
Design Guidelines ◽  
Selection System ◽  
Genetic Operators ◽  
Research Results ◽  
Parameter Combination ◽  
Determinant Factor

Performance of genetic algorithms is affected not only by each genetic operator, but also by the interaction among genetic operators. Research on this issue still fails to converge to any conclusion. In this paper, the author focuses mainly on investigating, through a series of systematic experiments, the effects of different combinations of parameter settings for genetic operators on the performance of the author’s GA-based batch selection system, and compare the research results with the claims made by previous research. One of the major findings of the author’s research is that the crossover rate is not as a determinant factor as the population size or the mutation rate in affecting a GA’s performance. This paper intends to serve as an inquiry into the research of useful design guidelines for parameterizing GA-based systems.

Asymptotic convergence of genetic algorithms

1998 ◽  
Vol 30 (2) ◽  
pp. 521-550 ◽  
Author(s):  
Raphaël Cerf
Keyword(s):  
Genetic Algorithms ◽  
Population Size ◽  
Fitness Function ◽  
Evolutionary Process ◽  
Asymptotic Convergence ◽  
Homogeneous Case ◽  
Selection Scheme ◽  
Convergence Results ◽  
Asymptotic Dynamics ◽  
Simple Selection

We study a markovian evolutionary process which encompasses the classical simple genetic algorithm. This process is obtained by randomly perturbing a very simple selection scheme. Using the Freidlin-Wentzell theory, we carry out a precise study of the asymptotic dynamics of the process as the perturbations disappear. We show how a delicate interaction between the perturbations and the selection pressure may force the convergence toward the global maxima of the fitness function. We put forward the existence of a critical population size, above which this kind of convergence can be achieved. We compute upper bounds of this critical population size for several examples. We derive several conditions to ensure convergence in the homogeneous case and these provide the first mathematically well-founded convergence results for genetic algorithms.

Sign in / Sign up

Export Citation Format

Share Document