SlideShare a Scribd company logo
1 of 9
Download to read offline
Research Inventy: International Journal Of Engineering And Science
Vol.4, Issue 3(March 2014), PP 46-54
Issn (e): 2278-4721, Issn (p):2319-6483, www.researchinventy.com
46
The Genetic Algorithm for finding the maxima of single-variable
functions
1
Dezdemona Gjylapi, 2
Vladimir Kasëmi
1
(Informatics Engineering Department, Pavaresia University, Vlore, Albania)
2
(Naval Engineering Department, University of Vlore, Albania)
ABSTRACT - The aim of this paper is to present an application of genetic algorithms for finding the maxima
of single variable functions, which is being tested on complex, multi-modal, non-differentiable and not continues
functions. Genetic algorithm using binary presentation of solutions, ranking selection and elitist selection, as
well as one point crossover and mutation for each bit, in fact concludes that if the function has only a maximum
all strings will converge to that value, but if the function is sufficient complex the algorithm finds a maxima very
close to the real maxima of the function, in a reasonable number of generations.
Keywords - Genetic algorithms, maxima, multimodal function, solution, stochastic sampling
I. INTRODUCTION
The algorithms for finding the functions’ maxima being typically a part of the optimization are
generally limited for regular convex functions. However, many functions are multimodal, not continuous, and
not-differentiable. Methods with stochastic selection are used to optimize these functions. While traditional
searching techniques use the characteristics of the problem to define the next point of the chosen sampling (e.g.
the gradient, hessian, linearity and continuity), the stochastic technique does not require such information,
except the value of the function at the points being analyzed. Instead, the other sampling points are defined
based more on the stochastic rules of selection or decision making than on several deterministic rules. Genetic
algorithms are used to solve difficult problems that have as objective functions that do not possess “good”
features such as continuity, derivability, fulfillment of Lipchitz conditions, etc. [1],[3],[4],[5].
The aim of this paper is to show how Genetic Algorithms (GA) is used for finding the maximum of
various functions, especially those that are complex, multimodal, non-differentiable and not continuous.
Section 2 will present the basics of GAs, in section 3 GA is tested over different multimodal continuous single-
variable functions and not continuous single-variable functions to show that GA is efficient in the optimization
process.
II. GENETIC ALGORITHMS OVERVIEW
A genetic algorithm applies the biological principles of natural evolution in the artificial systems. A
genetic algorithm is an iterative procedure that includes a population of individuals, each of them represented by
a finite string of symbols known as genes, which encode a possible solution in the space of a given problem.
This space, we refer to as the searching space, includes all the possible solutions of the respective problem. GA
is applied mainly in the spaces that are too big to be searched in a finite way. In Fig. 1 it is given a block scheme
which presents a simple GA.
The Genetic Algorithm in finding the maxima of single-variable functions
47
Figure 1. Block schema of a simple genetic algorithm
GAs starts with a population of individuals generated randomly, because GAs is a stochastic algorithm.
Randomness plays a fundamental role in GAs, either during selection phase as well as during reproduction.
Each individual of the population represents a possible solution of the problem that is being analyzed. The
individuals are evolved in the next iterations which are called generations. In each generation, the individuals
are assessed according to the fitness they have. The individuals that are selected to create new individuals
(offspring) are selected based on their fitness – the more adequate (fitting) they are, the more the chances they
have for reproduction.
Afterwards, the population of the next generation is created through the genetic operators. This
procedure continues until the termination criterion of the genetic algorithm is met (e.g. the number of the
populations or the improvement of the best solution).
The usage of one GA requires the definition of six fundamental issues: Solution representation,
creation of the initial population, evaluation function, termination criteria, selection function and genetic
operators. Genetic operators constitute the reproduction function to generate the new population. The rest of this
section we describes each of these topics.
A. Solution Representation
To better explain the steps of a genetic algorithm, we will examine a population of N individuals which
represent possible solutions to the problem. In GA, the individuals are represented as a string of length n like
below:
s= s1s2…sj...sn
The chromosome s is also called individual’s genotype, whilst the corresponding possible solution x of the string
s is called phenotype [6].
Usually, it is supposed that a one-to-one correspondence is established between genotypes and
phenotypes. However, depending on the situation, many-to-one and one-to-many relationships are also useful.
In both cases, the mapping from phenotype to genotype is called coding, and the mapping from genotype to
phenotype is called decoding [6].
Coding is the process of the representation of individuals (a.k.a chromosomes). The representation can
be using bits, numbers, trees, strings, lists or other objects. The coding depends mainly on the problem that
needs to be solved. The typical way of coding is the binary coding where each chromosome is coded as a binary
string.
Another possibility is that the whole string may represent a number and this is also the approach that
we have used for finding the functions’ maxima. The method of the representation by a binary string changes
from problem to problem. Binary string coded with 0 and 1, are the most used.
The Genetic Algorithm in finding the maxima of single-variable functions
48
B. Initialization and Termination
GA should generate a population to start with. In principle this can be a set of solutions generated in a
totally random manner. Certainly, the initial general fitness of these individuals is very bad. The advantage of a
population generated in such a manner is that it covers a big part of the solutions’ space. With the size increase
of the population, the biggest part of the solutions’ space can be covered, but the solution will be slower. It is
also possible to be covered only a given subset of the solution’s space, i.e.: if we know where to search for an
optimal solution.
The genetic algorithm passes from one generation to the other until the termination criteria of the
algorithm is met. The most used criteria for the termination of the algorithm is a fixed maximal number of
generations. There is another strategy uses the criteria of the population’s convergence. In general, GA pushes
the major part of the population to converge in one single solution. So, when the sum of deviations between
individuals becomes smaller than a predefined value, the algorithm can terminate. Also the algorithm may
terminate as result of the improvement of the best solution after a certain number of generations. Also, different
strategies can be combined together for terminating a genetic algorithm.
C. Evaluation function (fitness)
The nature obeys to Darwin’s principle “the survival of the strongest”, i.e., individuals with high fitness
value will be reproduces more often compared to those with low fitness value. In GA, the fitness is defined in
such a manner that the strongest strings have higher fitness values, and this is used to evaluate each individual of
a population. We should highlight here that fitness is the only relationship between GAs and the problem to be
solved, and this is the measure to select an individual to be reproduced for the next generation.
Likewise it is discussed in Goldberg [3], in the minimization’s problems, like the minimization of some costs’
functions z(x), introducing Cmax that satisfies the condition Cmax – z(x)>0, is preferable to be taken as a string’s
fitness function: f(x) = Cmax - z(x). Cmax can be taken as the greatest value z(x) in the current population, or as
the greatest value z(x) of t-generations.
Similarly, in maximization problems, such as maximizing the profit or the service, which are
represented as function u(x), if u(x) <0 for some values of x, introduces Cmin that satisfies u(x) + Cmin > 0, thus
the fitness function can be defined as f(x) = u (x) + Cmin. Cmin can be defined as the absolute value of the
smallest value of u(x) in the current population or after t-generations.
But, moreover, in the case of functions’ maximization we can use as fitness the function itself, and to
solve the problem in the case of its negative values, we use different selection methods.
D. Selection function
The selection of individuals to create the succeeding generations plays an extremely important role in a
genetic algorithm. Based on the individuals’ fitness, a random selection is made in such a way that the strongest
individuals have the biggest chances to be selected. An individual in one population can be selected more than
once and moreover, as the selection is random; all the individuals have the possibility to be selected in the next
generation. There exist some schemas for the selection process such as: the roulette and its versions, escalation
technique, ranking methods, tournament and elitism selection.
The usage of the roulette limits the genetic algorithm in the maximization process because the
evaluation function maps the solutions in a completely ordered set of values from R+.
In order to enable the minimization and the negativity in the fitness values, the roulette is completed with
elements such as windowing and scaling.
Ranking methods make possible that evaluation function maps the solutions in a partially ordered set of
values, thus enabling the process of minimization and negativity in the fitness values.
Ranking Selection means that only the range of individual fitness determines the probability of selection. In this
method the population is ranked from the best to the worst according to a criterion, which depends on the
problem [6].
The Genetic Algorithm in finding the maxima of single-variable functions
49
In the case of maximization of a function, the population will be ranked from the individual with the
biggest fitness to the one with the smaller fitness. The probability of the selection of each individual is
determined by its rank. There are several methods to determine the probability of selecting each individual
depending on whether the method is linear or not linear [6].
Likewise the ranking methods, the Tournament selection requires from the evaluation function only to
map the solutions to a partial ordered set. However, this order will not assign probabilities to the individuals.
Tournament selection works by randomly selecting j individuals from the current population, and introduces the
best one in the new population. This procedure is repeated until N individuals are selected [2].
Another genetic operator is also the Elitism, introduced for the first time by De Jong [7]. If the fitness
of an individual in the previous populations is greater than the fitness of each individual in current population,
elitism preserves the individual to the next generation.
E. Genetic Operators
Genetic operators provide the basic mechanism of searching the GA. Operators are used to create new
solutions based on existing solutions to the population. There are two basic types of operators: crossover and
mutation. Crossover takes two individuals and generates two new individuals, while mutation changes an
individual to generate a single new solution. The implementation of these two basic types of operators depends
on the method used for chromosome representation [2].
1) Crossover
The main distinctive feature of GA is the use of crossover. Crossover, also called recombination, is an
operator that creates new individuals from current population, combining pieces of information coming from
different individuals of the population. In fact, it recombines genetic material of two parent individuals to create
the offspring of the next generation [6].
Depending on how we are representing the individuals, various crossover techniques have been
created. When individuals are represented by binary strings {0.1}, some of the techniques used are single point
crossover, multipoint crossover and uniform crossover. Single point crossover, simple crossover otherwise,
randomly chooses a point in the strings of both parents, and then in order to create two offspring, the substrings
of the right side of the crossover point are exchanged.
In the multipoint crossover, several crossover points are randomly selected and in the uniform
crossover a random n-bits mask is chosen. Parity of each bit in the mask determines for each bit corresponding
to a seed, from which parent will get this bit. To be clear, for each bit position in the mask, i.e. its value "1" or
"0" indicates that the first parent or second parent contributes to the value of his first successor in that position,
and vice versa for the second successor.
The crossover is created with the hope that the new chromosomes will have the best part of the old
chromosomes, and maybe the new chromosomes will be better. However, it is best to leave a portion of the
population survive for the next generation.
Crossover Probability means how often the crossover will be performed. If the crossover probability is
100%, then the crossover creates all the offspring. More crossovers cause the calculation to be slower, but it will
have greater effect and will also increase the randomness. If the probability is 0%, whole new generation is
created from exact copies of chromosomes of the old population (but this does not mean that the new generation
is the same as the old one). Stopping the crossover is a bad idea; in this case, randomness is the only thing we
can work with [8].
2) Mutation
Mutation is a very easy method to create a new individual. This method can be used after the crossover
method. A mutation can be applied in several positions, once a new individual "is born". Mutation does a very
simple thing: it replaces one or more genes randomly selected in the individual’s representation.
Mutation probability means how often parts of chromosome will be subject to the mutation. If the
probability of mutation is 100%, the whole chromosome is changed, if it is 0%, nothing is changed.
Mutation is done to prevent the GA from falling into local optima, but should not happen very often,
because then the GA will actually turn into a random search and thus reduces efficiency [8].
The Genetic Algorithm in finding the maxima of single-variable functions
50
III. TESTING AND RESULTS
Testing is done with a single execution of the program with the following parameters:
i. The length of the gene: 10 digits after the decimal point, full part depends on the user.
ii. Size of population: 6
iii. Crossover: single point crossover
iv. Mutation: for each bit with probability 0.1
v. Selection method: Elitism and Linear Ranking using (1) for the assignment of probabilities.
21],
1
1
)1(2[
1
)( 


 
n
i
n
ip (1)
The application has been tested on examples of functions that present difficulties for finding their
maximum values. These examples are divided into two categories: continuous functions and not continuous
functions in their entire domain.
All the examples chosen, present difficulties in solving the equation obtained from equalization with
zero of the first order derivative of the function. In most cases are obtained equations of such orders, which have
no determined methods of solution.
In continuous functions are considered:
- Polynomial functions
- Neither Odd nor Even functions
- Rational functions with no asymptote
- Continuous transcendental functions
In not continuous functions are considered:
- Rational functions with vertical asymptote
- Rational functions with vertical and horizontal asymptote
- Transcendental functions with breaking points.
For all of the above functions are found their maximum values, through our application and to verify their
accuracy, corresponding graphs are built. In the example of Figure 12, which is a case that presents no difficulty
for finding its maximum value and in many other examples that do not appear in the study, the application has
found after a certain number of generations, the same maximum value (or with insignificant error) as standard
methods do.
A. Continuous functions
1. Polynomial neither Odd nor Even functions
Figure 2. Y=- x6
/60-x5
/50+x4
/2+2x3
/3-3.2x2
-6.4x
Figure 3. Fitness of Fig. 2 function depending on the
generation
The Genetic Algorithm in finding the maxima of single-variable functions
51
2. Rational functions with no asymptote
Figure 4. F(x) =-(x6
+x5
-10x2
-10x-12)/(x2
+6)
Figure 5. Fitness of Fig. 4 function depending on the
generation
3. Irrational functions
Figure 6. Y= 5-sqr(x4
+2x3
-3x2
+13)
Figure 7. Fitness of Fig. 6 function depending on the
generation
4. Trigonometric functions
Figure 8. Y=sin2
(3x+45) +0.9sin3
(9x)-sin (15x+50)
cos (2x-30)
Figure 9. Fitness of Fig.8 function depending on the
generation
The Genetic Algorithm in finding the maxima of single-variable functions
52
B. Not continuous functions
1. Rational functions with vertical asymptote
Figure 10. Y=sin (2πx)/x-cos (3πx)/x
Figure 11. Fitness of Fig. 10 function depending on
the generation
2. Rational functions with vertical and horizontal asymptote
Figure 12. F(x) = (2x-1)/x2
Figure 13. Fitness of Fig. 12 function depending on
the generation
3. Logarithmic functions
Figure 14. Y=- x4
(log |x| - 1) Figure 15. Fitness of the logarithmic function
depending on the generation
The Genetic Algorithm in finding the maxima of single-variable functions
53
4. Exponential functions
Figure 16. Y = e-x
(log |x|+x/|x|) Figure 17. Fitness of the exponential function
depending on the generation
Figure 18. Execution time of each function depending on the number of generations
Figure 19. Best, worst, and average execution time depending on number of generations
IV. CONCLUSION
GA can be used in optimization problems when the solution space grows very quickly, or an exact
solution is required in a limited time.
Implementation of the GA was tested successfully on the effectiveness and reliability for finding the
maximum of multimodal functions, not continuous, and complex, as a part of the optimization problems.
In the case of continuous multi-modal functions, after a certain number of generations the best solution does not
change.
The Genetic Algorithm in finding the maxima of single-variable functions
54
In the case of rational function with vertical asymptotes, which goes to infinity for x approaching 0
from the negative side of the coordinate axis, the accuracy of the solution depends on the accuracy of the
representation of numbers in the binary system. In this application accuracy used is 10 digits after the decimal
point in binary representation. With the increasing of the number of digits after the decimal point, the genetic
algorithm will find a better solution for this function.
In case of the functions that present no difficulty for finding their maximum value, the application has
found the same maximum value or with insignificant error, as standard methods do.
All the functions tested are multi-modal functions. So, by using nonlinear programming methods we
can take an optimal solution that varies depending on the initial values.
Whereas, despite the entire initial conditions, when problem is solved using GA, it will converge to the
true optimal value. This is influenced by the fact of multiple initial values to begin with, and the crossover and
mutation operations, to prevent trapping in local optimal solutions.
Even if a GA does not always give a perfect solution to a problem, it almost always can deliver at least
one good solution. In order to achieve this, the four main components of the the GA work together, i.e.:
parallelism, selection, mutation, and crossover.
REFERENCES
[1] Davis, Lawrence David. Handbook Of Genetic Algorithms (Van Nostrand Reinhold, 1991).
[2] A Genetic Algorithm for Function Optimization: A Matlab Implementation. Houck, Christopher R., Joines, Jeffery A. and Kay,
Michael G. 1998.
[3] Goldberg, D.E. Genetic Algorithms in Search, Optimization and Machine (Addison-Wesley, 1989).
[4] J.Holland. Adaptation in natural and artificial systems (The University of Michigan Press, 1975).
[5] Michalewicz, Z. Genetic Algorithms + Data Structures = Evolution Programs (New York : Springer-Verlag, 1994).
[6] SAKAWA, MASATOSHI. Genetic Algorithms and Fuzzy Multiobjective Optimization (Kluwer Academic Publishers, 2002). pp. 12-
26. ISBN 0-7923-7452-5.
[7] De Jong, K. A. An analysis of the behavior of a class of genetic adaptive systems (Ann Arbor,MI : University of Michigan, 1975).
[8] [Online]http://www.obitko.com/tutorials/genetic-algorithms/parameters.php

More Related Content

What's hot

Genetic algorithms in Data Mining
Genetic algorithms in Data MiningGenetic algorithms in Data Mining
Genetic algorithms in Data MiningAtul Khanna
 
Analysis of Parameter using Fuzzy Genetic Algorithm in E-learning System
Analysis of Parameter using Fuzzy Genetic Algorithm in E-learning SystemAnalysis of Parameter using Fuzzy Genetic Algorithm in E-learning System
Analysis of Parameter using Fuzzy Genetic Algorithm in E-learning SystemHarshal Jain
 
Fuzzy Genetic Algorithm
Fuzzy Genetic AlgorithmFuzzy Genetic Algorithm
Fuzzy Genetic AlgorithmPintu Khan
 
Genetic Algorithms - Artificial Intelligence
Genetic Algorithms - Artificial IntelligenceGenetic Algorithms - Artificial Intelligence
Genetic Algorithms - Artificial IntelligenceSahil Kumar
 
MACHINE LEARNING - GENETIC ALGORITHM
MACHINE LEARNING - GENETIC ALGORITHMMACHINE LEARNING - GENETIC ALGORITHM
MACHINE LEARNING - GENETIC ALGORITHMPuneet Kulyana
 
GENETIC ALGORITHM
GENETIC ALGORITHM GENETIC ALGORITHM
GENETIC ALGORITHM Abhishek Sur
 
Genetic algorithms
Genetic algorithmsGenetic algorithms
Genetic algorithmszamakhan
 
Genetic_Algorithm_AI(TU)
Genetic_Algorithm_AI(TU)Genetic_Algorithm_AI(TU)
Genetic_Algorithm_AI(TU)Kapil Khatiwada
 
Flowchart of GA
Flowchart of GAFlowchart of GA
Flowchart of GAIshucs
 
A New Method for Preserving Privacy in Data Publishing Against Attribute and ...
A New Method for Preserving Privacy in Data Publishing Against Attribute and ...A New Method for Preserving Privacy in Data Publishing Against Attribute and ...
A New Method for Preserving Privacy in Data Publishing Against Attribute and ...ijcisjournal
 

What's hot (19)

RM 701 Genetic Algorithm and Fuzzy Logic lecture
RM 701 Genetic Algorithm and Fuzzy Logic lectureRM 701 Genetic Algorithm and Fuzzy Logic lecture
RM 701 Genetic Algorithm and Fuzzy Logic lecture
 
Genetic algorithm
Genetic algorithmGenetic algorithm
Genetic algorithm
 
Genetic algorithms in Data Mining
Genetic algorithms in Data MiningGenetic algorithms in Data Mining
Genetic algorithms in Data Mining
 
Analysis of Parameter using Fuzzy Genetic Algorithm in E-learning System
Analysis of Parameter using Fuzzy Genetic Algorithm in E-learning SystemAnalysis of Parameter using Fuzzy Genetic Algorithm in E-learning System
Analysis of Parameter using Fuzzy Genetic Algorithm in E-learning System
 
Fuzzy Genetic Algorithm
Fuzzy Genetic AlgorithmFuzzy Genetic Algorithm
Fuzzy Genetic Algorithm
 
Genetic Algorithm
Genetic Algorithm Genetic Algorithm
Genetic Algorithm
 
Genetic Algorithms - Artificial Intelligence
Genetic Algorithms - Artificial IntelligenceGenetic Algorithms - Artificial Intelligence
Genetic Algorithms - Artificial Intelligence
 
MACHINE LEARNING - GENETIC ALGORITHM
MACHINE LEARNING - GENETIC ALGORITHMMACHINE LEARNING - GENETIC ALGORITHM
MACHINE LEARNING - GENETIC ALGORITHM
 
GENETIC ALGORITHM
GENETIC ALGORITHM GENETIC ALGORITHM
GENETIC ALGORITHM
 
Genetic algorithms
Genetic algorithmsGenetic algorithms
Genetic algorithms
 
Genetic_Algorithm_AI(TU)
Genetic_Algorithm_AI(TU)Genetic_Algorithm_AI(TU)
Genetic_Algorithm_AI(TU)
 
Flowchart of GA
Flowchart of GAFlowchart of GA
Flowchart of GA
 
Genetic Algorithm
Genetic AlgorithmGenetic Algorithm
Genetic Algorithm
 
Genetic Algorithms
Genetic AlgorithmsGenetic Algorithms
Genetic Algorithms
 
Ga
GaGa
Ga
 
Gadoc
GadocGadoc
Gadoc
 
F0422052058
F0422052058F0422052058
F0422052058
 
Genetic Algorithm
Genetic AlgorithmGenetic Algorithm
Genetic Algorithm
 
A New Method for Preserving Privacy in Data Publishing Against Attribute and ...
A New Method for Preserving Privacy in Data Publishing Against Attribute and ...A New Method for Preserving Privacy in Data Publishing Against Attribute and ...
A New Method for Preserving Privacy in Data Publishing Against Attribute and ...
 

Viewers also liked

Culture Shift in the Context of Language Shift
Culture Shift in the Context of Language ShiftCulture Shift in the Context of Language Shift
Culture Shift in the Context of Language Shiftinventy
 
A0420105
A0420105A0420105
A0420105inventy
 
Research Inventy : International Journal of Engineering and Science
Research Inventy : International Journal of Engineering and ScienceResearch Inventy : International Journal of Engineering and Science
Research Inventy : International Journal of Engineering and Scienceinventy
 
Novel Approach of Implementing Psychoacoustic model for MPEG-1 Audio
Novel Approach of Implementing Psychoacoustic model for MPEG-1 AudioNovel Approach of Implementing Psychoacoustic model for MPEG-1 Audio
Novel Approach of Implementing Psychoacoustic model for MPEG-1 Audioinventy
 
Research Inventy : International Journal of Engineering and Science
Research Inventy : International Journal of Engineering and ScienceResearch Inventy : International Journal of Engineering and Science
Research Inventy : International Journal of Engineering and Scienceinventy
 
Research Inventy : International Journal of Engineering and Science
Research Inventy : International Journal of Engineering and ScienceResearch Inventy : International Journal of Engineering and Science
Research Inventy : International Journal of Engineering and Scienceinventy
 
A generalized bernoulli sub-ODE Method and Its applications for nonlinear evo...
A generalized bernoulli sub-ODE Method and Its applications for nonlinear evo...A generalized bernoulli sub-ODE Method and Its applications for nonlinear evo...
A generalized bernoulli sub-ODE Method and Its applications for nonlinear evo...inventy
 
Research Inventy : International Journal of Engineering and Science
Research Inventy : International Journal of Engineering and ScienceResearch Inventy : International Journal of Engineering and Science
Research Inventy : International Journal of Engineering and Scienceinventy
 
Research Inventy : International Journal of Engineering and Science
Research Inventy : International Journal of Engineering and ScienceResearch Inventy : International Journal of Engineering and Science
Research Inventy : International Journal of Engineering and Scienceinventy
 

Viewers also liked (9)

Culture Shift in the Context of Language Shift
Culture Shift in the Context of Language ShiftCulture Shift in the Context of Language Shift
Culture Shift in the Context of Language Shift
 
A0420105
A0420105A0420105
A0420105
 
Research Inventy : International Journal of Engineering and Science
Research Inventy : International Journal of Engineering and ScienceResearch Inventy : International Journal of Engineering and Science
Research Inventy : International Journal of Engineering and Science
 
Novel Approach of Implementing Psychoacoustic model for MPEG-1 Audio
Novel Approach of Implementing Psychoacoustic model for MPEG-1 AudioNovel Approach of Implementing Psychoacoustic model for MPEG-1 Audio
Novel Approach of Implementing Psychoacoustic model for MPEG-1 Audio
 
Research Inventy : International Journal of Engineering and Science
Research Inventy : International Journal of Engineering and ScienceResearch Inventy : International Journal of Engineering and Science
Research Inventy : International Journal of Engineering and Science
 
Research Inventy : International Journal of Engineering and Science
Research Inventy : International Journal of Engineering and ScienceResearch Inventy : International Journal of Engineering and Science
Research Inventy : International Journal of Engineering and Science
 
A generalized bernoulli sub-ODE Method and Its applications for nonlinear evo...
A generalized bernoulli sub-ODE Method and Its applications for nonlinear evo...A generalized bernoulli sub-ODE Method and Its applications for nonlinear evo...
A generalized bernoulli sub-ODE Method and Its applications for nonlinear evo...
 
Research Inventy : International Journal of Engineering and Science
Research Inventy : International Journal of Engineering and ScienceResearch Inventy : International Journal of Engineering and Science
Research Inventy : International Journal of Engineering and Science
 
Research Inventy : International Journal of Engineering and Science
Research Inventy : International Journal of Engineering and ScienceResearch Inventy : International Journal of Engineering and Science
Research Inventy : International Journal of Engineering and Science
 

Similar to Genetic algorithm for finding maxima of functions

Genetic Algorithm 2 -.pptx
Genetic Algorithm 2 -.pptxGenetic Algorithm 2 -.pptx
Genetic Algorithm 2 -.pptxTAHANMKH
 
Genetic-Algorithms.ppt
Genetic-Algorithms.pptGenetic-Algorithms.ppt
Genetic-Algorithms.pptNipun85
 
AI_PPT_Genetic-Algorithms.ppt
AI_PPT_Genetic-Algorithms.pptAI_PPT_Genetic-Algorithms.ppt
AI_PPT_Genetic-Algorithms.pptHotTea
 
Genetic-Algorithms forv artificial .ppt
Genetic-Algorithms forv artificial  .pptGenetic-Algorithms forv artificial  .ppt
Genetic-Algorithms forv artificial .pptneelamsanjeevkumar
 
Genetic-Algorithms for machine learning and ai.ppt
Genetic-Algorithms for machine learning and ai.pptGenetic-Algorithms for machine learning and ai.ppt
Genetic-Algorithms for machine learning and ai.pptneelamsanjeevkumar
 
Genetic-Algorithms.ppt
Genetic-Algorithms.pptGenetic-Algorithms.ppt
Genetic-Algorithms.pptssuser2e437f
 
Genetic-Algorithms-computersciencepptnew.ppt
Genetic-Algorithms-computersciencepptnew.pptGenetic-Algorithms-computersciencepptnew.ppt
Genetic-Algorithms-computersciencepptnew.pptFitnessfreaksfam
 
Advanced Optimization Techniques
Advanced Optimization TechniquesAdvanced Optimization Techniques
Advanced Optimization TechniquesValerie Felton
 
Evolutionary computing - soft computing
Evolutionary computing - soft computingEvolutionary computing - soft computing
Evolutionary computing - soft computingSakshiMahto1
 
Genetic algorithm fitness function
Genetic algorithm fitness functionGenetic algorithm fitness function
Genetic algorithm fitness functionProf Ansari
 
4.Genetic-Algorithms.ppt
4.Genetic-Algorithms.ppt4.Genetic-Algorithms.ppt
4.Genetic-Algorithms.pptRamjiChaurasiya
 
Genetic algorithm
Genetic algorithmGenetic algorithm
Genetic algorithmgarima931
 
Multilevel techniques for the clustering problem
Multilevel techniques for the clustering problemMultilevel techniques for the clustering problem
Multilevel techniques for the clustering problemcsandit
 
soft computing BTU MCA 3rd SEM unit 1 .pptx
soft computing BTU MCA 3rd SEM unit 1 .pptxsoft computing BTU MCA 3rd SEM unit 1 .pptx
soft computing BTU MCA 3rd SEM unit 1 .pptxnaveen356604
 
A Genetic Algorithm Problem Solver For Archaeology
A Genetic Algorithm Problem Solver For ArchaeologyA Genetic Algorithm Problem Solver For Archaeology
A Genetic Algorithm Problem Solver For ArchaeologyAmy Cernava
 
Two-Stage Eagle Strategy with Differential Evolution
Two-Stage Eagle Strategy with Differential EvolutionTwo-Stage Eagle Strategy with Differential Evolution
Two-Stage Eagle Strategy with Differential EvolutionXin-She Yang
 
A Genetic Algorithm on Optimization Test Functions
A Genetic Algorithm on Optimization Test FunctionsA Genetic Algorithm on Optimization Test Functions
A Genetic Algorithm on Optimization Test FunctionsIJMERJOURNAL
 

Similar to Genetic algorithm for finding maxima of functions (20)

Genetic Algorithm 2 -.pptx
Genetic Algorithm 2 -.pptxGenetic Algorithm 2 -.pptx
Genetic Algorithm 2 -.pptx
 
Genetic-Algorithms.ppt
Genetic-Algorithms.pptGenetic-Algorithms.ppt
Genetic-Algorithms.ppt
 
AI_PPT_Genetic-Algorithms.ppt
AI_PPT_Genetic-Algorithms.pptAI_PPT_Genetic-Algorithms.ppt
AI_PPT_Genetic-Algorithms.ppt
 
Genetic-Algorithms forv artificial .ppt
Genetic-Algorithms forv artificial  .pptGenetic-Algorithms forv artificial  .ppt
Genetic-Algorithms forv artificial .ppt
 
Genetic-Algorithms for machine learning and ai.ppt
Genetic-Algorithms for machine learning and ai.pptGenetic-Algorithms for machine learning and ai.ppt
Genetic-Algorithms for machine learning and ai.ppt
 
Genetic-Algorithms.ppt
Genetic-Algorithms.pptGenetic-Algorithms.ppt
Genetic-Algorithms.ppt
 
Genetic-Algorithms-computersciencepptnew.ppt
Genetic-Algorithms-computersciencepptnew.pptGenetic-Algorithms-computersciencepptnew.ppt
Genetic-Algorithms-computersciencepptnew.ppt
 
Advanced Optimization Techniques
Advanced Optimization TechniquesAdvanced Optimization Techniques
Advanced Optimization Techniques
 
Evolutionary computing - soft computing
Evolutionary computing - soft computingEvolutionary computing - soft computing
Evolutionary computing - soft computing
 
L018147377
L018147377L018147377
L018147377
 
Genetic algorithm fitness function
Genetic algorithm fitness functionGenetic algorithm fitness function
Genetic algorithm fitness function
 
4.Genetic-Algorithms.ppt
4.Genetic-Algorithms.ppt4.Genetic-Algorithms.ppt
4.Genetic-Algorithms.ppt
 
Genetic algorithm
Genetic algorithmGenetic algorithm
Genetic algorithm
 
Multilevel techniques for the clustering problem
Multilevel techniques for the clustering problemMultilevel techniques for the clustering problem
Multilevel techniques for the clustering problem
 
soft computing BTU MCA 3rd SEM unit 1 .pptx
soft computing BTU MCA 3rd SEM unit 1 .pptxsoft computing BTU MCA 3rd SEM unit 1 .pptx
soft computing BTU MCA 3rd SEM unit 1 .pptx
 
A Genetic Algorithm Problem Solver For Archaeology
A Genetic Algorithm Problem Solver For ArchaeologyA Genetic Algorithm Problem Solver For Archaeology
A Genetic Algorithm Problem Solver For Archaeology
 
1582997627872.pdf
1582997627872.pdf1582997627872.pdf
1582997627872.pdf
 
Two-Stage Eagle Strategy with Differential Evolution
Two-Stage Eagle Strategy with Differential EvolutionTwo-Stage Eagle Strategy with Differential Evolution
Two-Stage Eagle Strategy with Differential Evolution
 
Genetic algorithms mahyar
Genetic algorithms   mahyarGenetic algorithms   mahyar
Genetic algorithms mahyar
 
A Genetic Algorithm on Optimization Test Functions
A Genetic Algorithm on Optimization Test FunctionsA Genetic Algorithm on Optimization Test Functions
A Genetic Algorithm on Optimization Test Functions
 

More from inventy

Experimental Investigation of a Household Refrigerator Using Evaporative-Cool...
Experimental Investigation of a Household Refrigerator Using Evaporative-Cool...Experimental Investigation of a Household Refrigerator Using Evaporative-Cool...
Experimental Investigation of a Household Refrigerator Using Evaporative-Cool...inventy
 
Copper Strip Corrossion Test in Various Aviation Fuels
Copper Strip Corrossion Test in Various Aviation FuelsCopper Strip Corrossion Test in Various Aviation Fuels
Copper Strip Corrossion Test in Various Aviation Fuelsinventy
 
Additional Conservation Laws for Two-Velocity Hydrodynamics Equations with th...
Additional Conservation Laws for Two-Velocity Hydrodynamics Equations with th...Additional Conservation Laws for Two-Velocity Hydrodynamics Equations with th...
Additional Conservation Laws for Two-Velocity Hydrodynamics Equations with th...inventy
 
Comparative Study of the Quality of Life, Quality of Work Life and Organisati...
Comparative Study of the Quality of Life, Quality of Work Life and Organisati...Comparative Study of the Quality of Life, Quality of Work Life and Organisati...
Comparative Study of the Quality of Life, Quality of Work Life and Organisati...inventy
 
A Study of Automated Decision Making Systems
A Study of Automated Decision Making SystemsA Study of Automated Decision Making Systems
A Study of Automated Decision Making Systemsinventy
 
Crystallization of L-Glutamic Acid: Mechanism of Heterogeneous β -Form Nuclea...
Crystallization of L-Glutamic Acid: Mechanism of Heterogeneous β -Form Nuclea...Crystallization of L-Glutamic Acid: Mechanism of Heterogeneous β -Form Nuclea...
Crystallization of L-Glutamic Acid: Mechanism of Heterogeneous β -Form Nuclea...inventy
 
Evaluation of Damage by the Reliability of the Traction Test on Polymer Test ...
Evaluation of Damage by the Reliability of the Traction Test on Polymer Test ...Evaluation of Damage by the Reliability of the Traction Test on Polymer Test ...
Evaluation of Damage by the Reliability of the Traction Test on Polymer Test ...inventy
 
Application of Kennelly’model of Running Performances to Elite Endurance Runn...
Application of Kennelly’model of Running Performances to Elite Endurance Runn...Application of Kennelly’model of Running Performances to Elite Endurance Runn...
Application of Kennelly’model of Running Performances to Elite Endurance Runn...inventy
 
Development and Application of a Failure Monitoring System by Using the Vibra...
Development and Application of a Failure Monitoring System by Using the Vibra...Development and Application of a Failure Monitoring System by Using the Vibra...
Development and Application of a Failure Monitoring System by Using the Vibra...inventy
 
The Management of Protected Areas in Serengeti Ecosystem: A Case Study of Iko...
The Management of Protected Areas in Serengeti Ecosystem: A Case Study of Iko...The Management of Protected Areas in Serengeti Ecosystem: A Case Study of Iko...
The Management of Protected Areas in Serengeti Ecosystem: A Case Study of Iko...inventy
 
Size distribution and biometric relationships of little tunny Euthynnus allet...
Size distribution and biometric relationships of little tunny Euthynnus allet...Size distribution and biometric relationships of little tunny Euthynnus allet...
Size distribution and biometric relationships of little tunny Euthynnus allet...inventy
 
Removal of Chromium (VI) From Aqueous Solutions Using Discarded Solanum Tuber...
Removal of Chromium (VI) From Aqueous Solutions Using Discarded Solanum Tuber...Removal of Chromium (VI) From Aqueous Solutions Using Discarded Solanum Tuber...
Removal of Chromium (VI) From Aqueous Solutions Using Discarded Solanum Tuber...inventy
 
Effect of Various External and Internal Factors on the Carrier Mobility in n-...
Effect of Various External and Internal Factors on the Carrier Mobility in n-...Effect of Various External and Internal Factors on the Carrier Mobility in n-...
Effect of Various External and Internal Factors on the Carrier Mobility in n-...inventy
 
Transient flow analysis for horizontal axial upper-wind turbine
Transient flow analysis for horizontal axial upper-wind turbineTransient flow analysis for horizontal axial upper-wind turbine
Transient flow analysis for horizontal axial upper-wind turbineinventy
 
Choice of Numerical Integration Method for Wind Time History Analysis of Tall...
Choice of Numerical Integration Method for Wind Time History Analysis of Tall...Choice of Numerical Integration Method for Wind Time History Analysis of Tall...
Choice of Numerical Integration Method for Wind Time History Analysis of Tall...inventy
 
Impacts of Demand Side Management on System Reliability Evaluation
Impacts of Demand Side Management on System Reliability EvaluationImpacts of Demand Side Management on System Reliability Evaluation
Impacts of Demand Side Management on System Reliability Evaluationinventy
 
Reliability Evaluation of Riyadh System Incorporating Renewable Generation
Reliability Evaluation of Riyadh System Incorporating Renewable GenerationReliability Evaluation of Riyadh System Incorporating Renewable Generation
Reliability Evaluation of Riyadh System Incorporating Renewable Generationinventy
 
The effect of reduced pressure acetylene plasma treatment on physical charact...
The effect of reduced pressure acetylene plasma treatment on physical charact...The effect of reduced pressure acetylene plasma treatment on physical charact...
The effect of reduced pressure acetylene plasma treatment on physical charact...inventy
 
Experimental Investigation of Mini Cooler cum Freezer
Experimental Investigation of Mini Cooler cum FreezerExperimental Investigation of Mini Cooler cum Freezer
Experimental Investigation of Mini Cooler cum Freezerinventy
 
Growth and Magnetic properties of MnGeP2 thin films
Growth and Magnetic properties of MnGeP2 thin filmsGrowth and Magnetic properties of MnGeP2 thin films
Growth and Magnetic properties of MnGeP2 thin filmsinventy
 

More from inventy (20)

Experimental Investigation of a Household Refrigerator Using Evaporative-Cool...
Experimental Investigation of a Household Refrigerator Using Evaporative-Cool...Experimental Investigation of a Household Refrigerator Using Evaporative-Cool...
Experimental Investigation of a Household Refrigerator Using Evaporative-Cool...
 
Copper Strip Corrossion Test in Various Aviation Fuels
Copper Strip Corrossion Test in Various Aviation FuelsCopper Strip Corrossion Test in Various Aviation Fuels
Copper Strip Corrossion Test in Various Aviation Fuels
 
Additional Conservation Laws for Two-Velocity Hydrodynamics Equations with th...
Additional Conservation Laws for Two-Velocity Hydrodynamics Equations with th...Additional Conservation Laws for Two-Velocity Hydrodynamics Equations with th...
Additional Conservation Laws for Two-Velocity Hydrodynamics Equations with th...
 
Comparative Study of the Quality of Life, Quality of Work Life and Organisati...
Comparative Study of the Quality of Life, Quality of Work Life and Organisati...Comparative Study of the Quality of Life, Quality of Work Life and Organisati...
Comparative Study of the Quality of Life, Quality of Work Life and Organisati...
 
A Study of Automated Decision Making Systems
A Study of Automated Decision Making SystemsA Study of Automated Decision Making Systems
A Study of Automated Decision Making Systems
 
Crystallization of L-Glutamic Acid: Mechanism of Heterogeneous β -Form Nuclea...
Crystallization of L-Glutamic Acid: Mechanism of Heterogeneous β -Form Nuclea...Crystallization of L-Glutamic Acid: Mechanism of Heterogeneous β -Form Nuclea...
Crystallization of L-Glutamic Acid: Mechanism of Heterogeneous β -Form Nuclea...
 
Evaluation of Damage by the Reliability of the Traction Test on Polymer Test ...
Evaluation of Damage by the Reliability of the Traction Test on Polymer Test ...Evaluation of Damage by the Reliability of the Traction Test on Polymer Test ...
Evaluation of Damage by the Reliability of the Traction Test on Polymer Test ...
 
Application of Kennelly’model of Running Performances to Elite Endurance Runn...
Application of Kennelly’model of Running Performances to Elite Endurance Runn...Application of Kennelly’model of Running Performances to Elite Endurance Runn...
Application of Kennelly’model of Running Performances to Elite Endurance Runn...
 
Development and Application of a Failure Monitoring System by Using the Vibra...
Development and Application of a Failure Monitoring System by Using the Vibra...Development and Application of a Failure Monitoring System by Using the Vibra...
Development and Application of a Failure Monitoring System by Using the Vibra...
 
The Management of Protected Areas in Serengeti Ecosystem: A Case Study of Iko...
The Management of Protected Areas in Serengeti Ecosystem: A Case Study of Iko...The Management of Protected Areas in Serengeti Ecosystem: A Case Study of Iko...
The Management of Protected Areas in Serengeti Ecosystem: A Case Study of Iko...
 
Size distribution and biometric relationships of little tunny Euthynnus allet...
Size distribution and biometric relationships of little tunny Euthynnus allet...Size distribution and biometric relationships of little tunny Euthynnus allet...
Size distribution and biometric relationships of little tunny Euthynnus allet...
 
Removal of Chromium (VI) From Aqueous Solutions Using Discarded Solanum Tuber...
Removal of Chromium (VI) From Aqueous Solutions Using Discarded Solanum Tuber...Removal of Chromium (VI) From Aqueous Solutions Using Discarded Solanum Tuber...
Removal of Chromium (VI) From Aqueous Solutions Using Discarded Solanum Tuber...
 
Effect of Various External and Internal Factors on the Carrier Mobility in n-...
Effect of Various External and Internal Factors on the Carrier Mobility in n-...Effect of Various External and Internal Factors on the Carrier Mobility in n-...
Effect of Various External and Internal Factors on the Carrier Mobility in n-...
 
Transient flow analysis for horizontal axial upper-wind turbine
Transient flow analysis for horizontal axial upper-wind turbineTransient flow analysis for horizontal axial upper-wind turbine
Transient flow analysis for horizontal axial upper-wind turbine
 
Choice of Numerical Integration Method for Wind Time History Analysis of Tall...
Choice of Numerical Integration Method for Wind Time History Analysis of Tall...Choice of Numerical Integration Method for Wind Time History Analysis of Tall...
Choice of Numerical Integration Method for Wind Time History Analysis of Tall...
 
Impacts of Demand Side Management on System Reliability Evaluation
Impacts of Demand Side Management on System Reliability EvaluationImpacts of Demand Side Management on System Reliability Evaluation
Impacts of Demand Side Management on System Reliability Evaluation
 
Reliability Evaluation of Riyadh System Incorporating Renewable Generation
Reliability Evaluation of Riyadh System Incorporating Renewable GenerationReliability Evaluation of Riyadh System Incorporating Renewable Generation
Reliability Evaluation of Riyadh System Incorporating Renewable Generation
 
The effect of reduced pressure acetylene plasma treatment on physical charact...
The effect of reduced pressure acetylene plasma treatment on physical charact...The effect of reduced pressure acetylene plasma treatment on physical charact...
The effect of reduced pressure acetylene plasma treatment on physical charact...
 
Experimental Investigation of Mini Cooler cum Freezer
Experimental Investigation of Mini Cooler cum FreezerExperimental Investigation of Mini Cooler cum Freezer
Experimental Investigation of Mini Cooler cum Freezer
 
Growth and Magnetic properties of MnGeP2 thin films
Growth and Magnetic properties of MnGeP2 thin filmsGrowth and Magnetic properties of MnGeP2 thin films
Growth and Magnetic properties of MnGeP2 thin films
 

Recently uploaded

08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking MenDelhi Call girls
 
Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024Scott Keck-Warren
 
Next-generation AAM aircraft unveiled by Supernal, S-A2
Next-generation AAM aircraft unveiled by Supernal, S-A2Next-generation AAM aircraft unveiled by Supernal, S-A2
Next-generation AAM aircraft unveiled by Supernal, S-A2Hyundai Motor Group
 
Hyderabad Call Girls Khairatabad ✨ 7001305949 ✨ Cheap Price Your Budget
Hyderabad Call Girls Khairatabad ✨ 7001305949 ✨ Cheap Price Your BudgetHyderabad Call Girls Khairatabad ✨ 7001305949 ✨ Cheap Price Your Budget
Hyderabad Call Girls Khairatabad ✨ 7001305949 ✨ Cheap Price Your BudgetEnjoy Anytime
 
Artificial intelligence in the post-deep learning era
Artificial intelligence in the post-deep learning eraArtificial intelligence in the post-deep learning era
Artificial intelligence in the post-deep learning eraDeakin University
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking MenDelhi Call girls
 
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Patryk Bandurski
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationSafe Software
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreternaman860154
 
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Alan Dix
 
Pigging Solutions in Pet Food Manufacturing
Pigging Solutions in Pet Food ManufacturingPigging Solutions in Pet Food Manufacturing
Pigging Solutions in Pet Food ManufacturingPigging Solutions
 
AI as an Interface for Commercial Buildings
AI as an Interface for Commercial BuildingsAI as an Interface for Commercial Buildings
AI as an Interface for Commercial BuildingsMemoori
 
Snow Chain-Integrated Tire for a Safe Drive on Winter Roads
Snow Chain-Integrated Tire for a Safe Drive on Winter RoadsSnow Chain-Integrated Tire for a Safe Drive on Winter Roads
Snow Chain-Integrated Tire for a Safe Drive on Winter RoadsHyundai Motor Group
 
Azure Monitor & Application Insight to monitor Infrastructure & Application
Azure Monitor & Application Insight to monitor Infrastructure & ApplicationAzure Monitor & Application Insight to monitor Infrastructure & Application
Azure Monitor & Application Insight to monitor Infrastructure & ApplicationAndikSusilo4
 
How to convert PDF to text with Nanonets
How to convert PDF to text with NanonetsHow to convert PDF to text with Nanonets
How to convert PDF to text with Nanonetsnaman860154
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...shyamraj55
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsMark Billinghurst
 
Injustice - Developers Among Us (SciFiDevCon 2024)
Injustice - Developers Among Us (SciFiDevCon 2024)Injustice - Developers Among Us (SciFiDevCon 2024)
Injustice - Developers Among Us (SciFiDevCon 2024)Allon Mureinik
 
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...HostedbyConfluent
 

Recently uploaded (20)

08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
 
Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024
 
Next-generation AAM aircraft unveiled by Supernal, S-A2
Next-generation AAM aircraft unveiled by Supernal, S-A2Next-generation AAM aircraft unveiled by Supernal, S-A2
Next-generation AAM aircraft unveiled by Supernal, S-A2
 
Hyderabad Call Girls Khairatabad ✨ 7001305949 ✨ Cheap Price Your Budget
Hyderabad Call Girls Khairatabad ✨ 7001305949 ✨ Cheap Price Your BudgetHyderabad Call Girls Khairatabad ✨ 7001305949 ✨ Cheap Price Your Budget
Hyderabad Call Girls Khairatabad ✨ 7001305949 ✨ Cheap Price Your Budget
 
Artificial intelligence in the post-deep learning era
Artificial intelligence in the post-deep learning eraArtificial intelligence in the post-deep learning era
Artificial intelligence in the post-deep learning era
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men
 
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreter
 
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
 
Pigging Solutions in Pet Food Manufacturing
Pigging Solutions in Pet Food ManufacturingPigging Solutions in Pet Food Manufacturing
Pigging Solutions in Pet Food Manufacturing
 
AI as an Interface for Commercial Buildings
AI as an Interface for Commercial BuildingsAI as an Interface for Commercial Buildings
AI as an Interface for Commercial Buildings
 
Vulnerability_Management_GRC_by Sohang Sengupta.pptx
Vulnerability_Management_GRC_by Sohang Sengupta.pptxVulnerability_Management_GRC_by Sohang Sengupta.pptx
Vulnerability_Management_GRC_by Sohang Sengupta.pptx
 
Snow Chain-Integrated Tire for a Safe Drive on Winter Roads
Snow Chain-Integrated Tire for a Safe Drive on Winter RoadsSnow Chain-Integrated Tire for a Safe Drive on Winter Roads
Snow Chain-Integrated Tire for a Safe Drive on Winter Roads
 
Azure Monitor & Application Insight to monitor Infrastructure & Application
Azure Monitor & Application Insight to monitor Infrastructure & ApplicationAzure Monitor & Application Insight to monitor Infrastructure & Application
Azure Monitor & Application Insight to monitor Infrastructure & Application
 
How to convert PDF to text with Nanonets
How to convert PDF to text with NanonetsHow to convert PDF to text with Nanonets
How to convert PDF to text with Nanonets
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR Systems
 
Injustice - Developers Among Us (SciFiDevCon 2024)
Injustice - Developers Among Us (SciFiDevCon 2024)Injustice - Developers Among Us (SciFiDevCon 2024)
Injustice - Developers Among Us (SciFiDevCon 2024)
 
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
 

Genetic algorithm for finding maxima of functions

  • 1. Research Inventy: International Journal Of Engineering And Science Vol.4, Issue 3(March 2014), PP 46-54 Issn (e): 2278-4721, Issn (p):2319-6483, www.researchinventy.com 46 The Genetic Algorithm for finding the maxima of single-variable functions 1 Dezdemona Gjylapi, 2 Vladimir Kasëmi 1 (Informatics Engineering Department, Pavaresia University, Vlore, Albania) 2 (Naval Engineering Department, University of Vlore, Albania) ABSTRACT - The aim of this paper is to present an application of genetic algorithms for finding the maxima of single variable functions, which is being tested on complex, multi-modal, non-differentiable and not continues functions. Genetic algorithm using binary presentation of solutions, ranking selection and elitist selection, as well as one point crossover and mutation for each bit, in fact concludes that if the function has only a maximum all strings will converge to that value, but if the function is sufficient complex the algorithm finds a maxima very close to the real maxima of the function, in a reasonable number of generations. Keywords - Genetic algorithms, maxima, multimodal function, solution, stochastic sampling I. INTRODUCTION The algorithms for finding the functions’ maxima being typically a part of the optimization are generally limited for regular convex functions. However, many functions are multimodal, not continuous, and not-differentiable. Methods with stochastic selection are used to optimize these functions. While traditional searching techniques use the characteristics of the problem to define the next point of the chosen sampling (e.g. the gradient, hessian, linearity and continuity), the stochastic technique does not require such information, except the value of the function at the points being analyzed. Instead, the other sampling points are defined based more on the stochastic rules of selection or decision making than on several deterministic rules. Genetic algorithms are used to solve difficult problems that have as objective functions that do not possess “good” features such as continuity, derivability, fulfillment of Lipchitz conditions, etc. [1],[3],[4],[5]. The aim of this paper is to show how Genetic Algorithms (GA) is used for finding the maximum of various functions, especially those that are complex, multimodal, non-differentiable and not continuous. Section 2 will present the basics of GAs, in section 3 GA is tested over different multimodal continuous single- variable functions and not continuous single-variable functions to show that GA is efficient in the optimization process. II. GENETIC ALGORITHMS OVERVIEW A genetic algorithm applies the biological principles of natural evolution in the artificial systems. A genetic algorithm is an iterative procedure that includes a population of individuals, each of them represented by a finite string of symbols known as genes, which encode a possible solution in the space of a given problem. This space, we refer to as the searching space, includes all the possible solutions of the respective problem. GA is applied mainly in the spaces that are too big to be searched in a finite way. In Fig. 1 it is given a block scheme which presents a simple GA.
  • 2. The Genetic Algorithm in finding the maxima of single-variable functions 47 Figure 1. Block schema of a simple genetic algorithm GAs starts with a population of individuals generated randomly, because GAs is a stochastic algorithm. Randomness plays a fundamental role in GAs, either during selection phase as well as during reproduction. Each individual of the population represents a possible solution of the problem that is being analyzed. The individuals are evolved in the next iterations which are called generations. In each generation, the individuals are assessed according to the fitness they have. The individuals that are selected to create new individuals (offspring) are selected based on their fitness – the more adequate (fitting) they are, the more the chances they have for reproduction. Afterwards, the population of the next generation is created through the genetic operators. This procedure continues until the termination criterion of the genetic algorithm is met (e.g. the number of the populations or the improvement of the best solution). The usage of one GA requires the definition of six fundamental issues: Solution representation, creation of the initial population, evaluation function, termination criteria, selection function and genetic operators. Genetic operators constitute the reproduction function to generate the new population. The rest of this section we describes each of these topics. A. Solution Representation To better explain the steps of a genetic algorithm, we will examine a population of N individuals which represent possible solutions to the problem. In GA, the individuals are represented as a string of length n like below: s= s1s2…sj...sn The chromosome s is also called individual’s genotype, whilst the corresponding possible solution x of the string s is called phenotype [6]. Usually, it is supposed that a one-to-one correspondence is established between genotypes and phenotypes. However, depending on the situation, many-to-one and one-to-many relationships are also useful. In both cases, the mapping from phenotype to genotype is called coding, and the mapping from genotype to phenotype is called decoding [6]. Coding is the process of the representation of individuals (a.k.a chromosomes). The representation can be using bits, numbers, trees, strings, lists or other objects. The coding depends mainly on the problem that needs to be solved. The typical way of coding is the binary coding where each chromosome is coded as a binary string. Another possibility is that the whole string may represent a number and this is also the approach that we have used for finding the functions’ maxima. The method of the representation by a binary string changes from problem to problem. Binary string coded with 0 and 1, are the most used.
  • 3. The Genetic Algorithm in finding the maxima of single-variable functions 48 B. Initialization and Termination GA should generate a population to start with. In principle this can be a set of solutions generated in a totally random manner. Certainly, the initial general fitness of these individuals is very bad. The advantage of a population generated in such a manner is that it covers a big part of the solutions’ space. With the size increase of the population, the biggest part of the solutions’ space can be covered, but the solution will be slower. It is also possible to be covered only a given subset of the solution’s space, i.e.: if we know where to search for an optimal solution. The genetic algorithm passes from one generation to the other until the termination criteria of the algorithm is met. The most used criteria for the termination of the algorithm is a fixed maximal number of generations. There is another strategy uses the criteria of the population’s convergence. In general, GA pushes the major part of the population to converge in one single solution. So, when the sum of deviations between individuals becomes smaller than a predefined value, the algorithm can terminate. Also the algorithm may terminate as result of the improvement of the best solution after a certain number of generations. Also, different strategies can be combined together for terminating a genetic algorithm. C. Evaluation function (fitness) The nature obeys to Darwin’s principle “the survival of the strongest”, i.e., individuals with high fitness value will be reproduces more often compared to those with low fitness value. In GA, the fitness is defined in such a manner that the strongest strings have higher fitness values, and this is used to evaluate each individual of a population. We should highlight here that fitness is the only relationship between GAs and the problem to be solved, and this is the measure to select an individual to be reproduced for the next generation. Likewise it is discussed in Goldberg [3], in the minimization’s problems, like the minimization of some costs’ functions z(x), introducing Cmax that satisfies the condition Cmax – z(x)>0, is preferable to be taken as a string’s fitness function: f(x) = Cmax - z(x). Cmax can be taken as the greatest value z(x) in the current population, or as the greatest value z(x) of t-generations. Similarly, in maximization problems, such as maximizing the profit or the service, which are represented as function u(x), if u(x) <0 for some values of x, introduces Cmin that satisfies u(x) + Cmin > 0, thus the fitness function can be defined as f(x) = u (x) + Cmin. Cmin can be defined as the absolute value of the smallest value of u(x) in the current population or after t-generations. But, moreover, in the case of functions’ maximization we can use as fitness the function itself, and to solve the problem in the case of its negative values, we use different selection methods. D. Selection function The selection of individuals to create the succeeding generations plays an extremely important role in a genetic algorithm. Based on the individuals’ fitness, a random selection is made in such a way that the strongest individuals have the biggest chances to be selected. An individual in one population can be selected more than once and moreover, as the selection is random; all the individuals have the possibility to be selected in the next generation. There exist some schemas for the selection process such as: the roulette and its versions, escalation technique, ranking methods, tournament and elitism selection. The usage of the roulette limits the genetic algorithm in the maximization process because the evaluation function maps the solutions in a completely ordered set of values from R+. In order to enable the minimization and the negativity in the fitness values, the roulette is completed with elements such as windowing and scaling. Ranking methods make possible that evaluation function maps the solutions in a partially ordered set of values, thus enabling the process of minimization and negativity in the fitness values. Ranking Selection means that only the range of individual fitness determines the probability of selection. In this method the population is ranked from the best to the worst according to a criterion, which depends on the problem [6].
  • 4. The Genetic Algorithm in finding the maxima of single-variable functions 49 In the case of maximization of a function, the population will be ranked from the individual with the biggest fitness to the one with the smaller fitness. The probability of the selection of each individual is determined by its rank. There are several methods to determine the probability of selecting each individual depending on whether the method is linear or not linear [6]. Likewise the ranking methods, the Tournament selection requires from the evaluation function only to map the solutions to a partial ordered set. However, this order will not assign probabilities to the individuals. Tournament selection works by randomly selecting j individuals from the current population, and introduces the best one in the new population. This procedure is repeated until N individuals are selected [2]. Another genetic operator is also the Elitism, introduced for the first time by De Jong [7]. If the fitness of an individual in the previous populations is greater than the fitness of each individual in current population, elitism preserves the individual to the next generation. E. Genetic Operators Genetic operators provide the basic mechanism of searching the GA. Operators are used to create new solutions based on existing solutions to the population. There are two basic types of operators: crossover and mutation. Crossover takes two individuals and generates two new individuals, while mutation changes an individual to generate a single new solution. The implementation of these two basic types of operators depends on the method used for chromosome representation [2]. 1) Crossover The main distinctive feature of GA is the use of crossover. Crossover, also called recombination, is an operator that creates new individuals from current population, combining pieces of information coming from different individuals of the population. In fact, it recombines genetic material of two parent individuals to create the offspring of the next generation [6]. Depending on how we are representing the individuals, various crossover techniques have been created. When individuals are represented by binary strings {0.1}, some of the techniques used are single point crossover, multipoint crossover and uniform crossover. Single point crossover, simple crossover otherwise, randomly chooses a point in the strings of both parents, and then in order to create two offspring, the substrings of the right side of the crossover point are exchanged. In the multipoint crossover, several crossover points are randomly selected and in the uniform crossover a random n-bits mask is chosen. Parity of each bit in the mask determines for each bit corresponding to a seed, from which parent will get this bit. To be clear, for each bit position in the mask, i.e. its value "1" or "0" indicates that the first parent or second parent contributes to the value of his first successor in that position, and vice versa for the second successor. The crossover is created with the hope that the new chromosomes will have the best part of the old chromosomes, and maybe the new chromosomes will be better. However, it is best to leave a portion of the population survive for the next generation. Crossover Probability means how often the crossover will be performed. If the crossover probability is 100%, then the crossover creates all the offspring. More crossovers cause the calculation to be slower, but it will have greater effect and will also increase the randomness. If the probability is 0%, whole new generation is created from exact copies of chromosomes of the old population (but this does not mean that the new generation is the same as the old one). Stopping the crossover is a bad idea; in this case, randomness is the only thing we can work with [8]. 2) Mutation Mutation is a very easy method to create a new individual. This method can be used after the crossover method. A mutation can be applied in several positions, once a new individual "is born". Mutation does a very simple thing: it replaces one or more genes randomly selected in the individual’s representation. Mutation probability means how often parts of chromosome will be subject to the mutation. If the probability of mutation is 100%, the whole chromosome is changed, if it is 0%, nothing is changed. Mutation is done to prevent the GA from falling into local optima, but should not happen very often, because then the GA will actually turn into a random search and thus reduces efficiency [8].
  • 5. The Genetic Algorithm in finding the maxima of single-variable functions 50 III. TESTING AND RESULTS Testing is done with a single execution of the program with the following parameters: i. The length of the gene: 10 digits after the decimal point, full part depends on the user. ii. Size of population: 6 iii. Crossover: single point crossover iv. Mutation: for each bit with probability 0.1 v. Selection method: Elitism and Linear Ranking using (1) for the assignment of probabilities. 21], 1 1 )1(2[ 1 )(      n i n ip (1) The application has been tested on examples of functions that present difficulties for finding their maximum values. These examples are divided into two categories: continuous functions and not continuous functions in their entire domain. All the examples chosen, present difficulties in solving the equation obtained from equalization with zero of the first order derivative of the function. In most cases are obtained equations of such orders, which have no determined methods of solution. In continuous functions are considered: - Polynomial functions - Neither Odd nor Even functions - Rational functions with no asymptote - Continuous transcendental functions In not continuous functions are considered: - Rational functions with vertical asymptote - Rational functions with vertical and horizontal asymptote - Transcendental functions with breaking points. For all of the above functions are found their maximum values, through our application and to verify their accuracy, corresponding graphs are built. In the example of Figure 12, which is a case that presents no difficulty for finding its maximum value and in many other examples that do not appear in the study, the application has found after a certain number of generations, the same maximum value (or with insignificant error) as standard methods do. A. Continuous functions 1. Polynomial neither Odd nor Even functions Figure 2. Y=- x6 /60-x5 /50+x4 /2+2x3 /3-3.2x2 -6.4x Figure 3. Fitness of Fig. 2 function depending on the generation
  • 6. The Genetic Algorithm in finding the maxima of single-variable functions 51 2. Rational functions with no asymptote Figure 4. F(x) =-(x6 +x5 -10x2 -10x-12)/(x2 +6) Figure 5. Fitness of Fig. 4 function depending on the generation 3. Irrational functions Figure 6. Y= 5-sqr(x4 +2x3 -3x2 +13) Figure 7. Fitness of Fig. 6 function depending on the generation 4. Trigonometric functions Figure 8. Y=sin2 (3x+45) +0.9sin3 (9x)-sin (15x+50) cos (2x-30) Figure 9. Fitness of Fig.8 function depending on the generation
  • 7. The Genetic Algorithm in finding the maxima of single-variable functions 52 B. Not continuous functions 1. Rational functions with vertical asymptote Figure 10. Y=sin (2πx)/x-cos (3πx)/x Figure 11. Fitness of Fig. 10 function depending on the generation 2. Rational functions with vertical and horizontal asymptote Figure 12. F(x) = (2x-1)/x2 Figure 13. Fitness of Fig. 12 function depending on the generation 3. Logarithmic functions Figure 14. Y=- x4 (log |x| - 1) Figure 15. Fitness of the logarithmic function depending on the generation
  • 8. The Genetic Algorithm in finding the maxima of single-variable functions 53 4. Exponential functions Figure 16. Y = e-x (log |x|+x/|x|) Figure 17. Fitness of the exponential function depending on the generation Figure 18. Execution time of each function depending on the number of generations Figure 19. Best, worst, and average execution time depending on number of generations IV. CONCLUSION GA can be used in optimization problems when the solution space grows very quickly, or an exact solution is required in a limited time. Implementation of the GA was tested successfully on the effectiveness and reliability for finding the maximum of multimodal functions, not continuous, and complex, as a part of the optimization problems. In the case of continuous multi-modal functions, after a certain number of generations the best solution does not change.
  • 9. The Genetic Algorithm in finding the maxima of single-variable functions 54 In the case of rational function with vertical asymptotes, which goes to infinity for x approaching 0 from the negative side of the coordinate axis, the accuracy of the solution depends on the accuracy of the representation of numbers in the binary system. In this application accuracy used is 10 digits after the decimal point in binary representation. With the increasing of the number of digits after the decimal point, the genetic algorithm will find a better solution for this function. In case of the functions that present no difficulty for finding their maximum value, the application has found the same maximum value or with insignificant error, as standard methods do. All the functions tested are multi-modal functions. So, by using nonlinear programming methods we can take an optimal solution that varies depending on the initial values. Whereas, despite the entire initial conditions, when problem is solved using GA, it will converge to the true optimal value. This is influenced by the fact of multiple initial values to begin with, and the crossover and mutation operations, to prevent trapping in local optimal solutions. Even if a GA does not always give a perfect solution to a problem, it almost always can deliver at least one good solution. In order to achieve this, the four main components of the the GA work together, i.e.: parallelism, selection, mutation, and crossover. REFERENCES [1] Davis, Lawrence David. Handbook Of Genetic Algorithms (Van Nostrand Reinhold, 1991). [2] A Genetic Algorithm for Function Optimization: A Matlab Implementation. Houck, Christopher R., Joines, Jeffery A. and Kay, Michael G. 1998. [3] Goldberg, D.E. Genetic Algorithms in Search, Optimization and Machine (Addison-Wesley, 1989). [4] J.Holland. Adaptation in natural and artificial systems (The University of Michigan Press, 1975). [5] Michalewicz, Z. Genetic Algorithms + Data Structures = Evolution Programs (New York : Springer-Verlag, 1994). [6] SAKAWA, MASATOSHI. Genetic Algorithms and Fuzzy Multiobjective Optimization (Kluwer Academic Publishers, 2002). pp. 12- 26. ISBN 0-7923-7452-5. [7] De Jong, K. A. An analysis of the behavior of a class of genetic adaptive systems (Ann Arbor,MI : University of Michigan, 1975). [8] [Online]http://www.obitko.com/tutorials/genetic-algorithms/parameters.php