13
GENETIC ALGORITHMS Ranga Rodrigo March 5, 2014 1

Genetic Algorithms

  • Upload
    ugo

  • View
    43

  • Download
    0

Embed Size (px)

DESCRIPTION

Genetic Algorithms. Ranga Rodrigo March 5, 2014. Evolutionary Computation (EC). Introduction to Evolutionary Computation. Evolution is this process of adaption with the aim of improving the survival capabilities through processes such as natural selection, survival of the fittest, - PowerPoint PPT Presentation

Citation preview

Page 1: Genetic Algorithms

1

GENETIC ALGORITHMS

Ranga RodrigoMarch 5, 2014

Page 2: Genetic Algorithms

2

EVOLUTIONARY COMPUTATION (EC)

Page 3: Genetic Algorithms

3

INTRODUCTION TO EVOLUTIONARY COMPUTATION

• Evolution is this process of adaption with the aim of improving the survival capabilities through processes such as – natural selection, – survival of the fittest, – reproduction, – mutation, – competition and – symbiosis.

Page 4: Genetic Algorithms

4

DNA, the molecular basis for inheritance. Each strand of DNA is a chain of nucleotides, matching each other in the center to form what look like rungs on a twisted ladder.

http://en.wikipedia.org/wiki/Genetics

Page 5: Genetic Algorithms

5

A Punnett square depicting a cross between two pea plants heterozygous for purple (B) and white (b) blossoms. At its most fundamental level, inheritance in organisms occurs by passing discrete heritable units, called genes, from parents to progeny.[31] This property was first observed by Gregor Mendel, who studied the segregation of heritable traits in pea plants.[12][32] In his experiments studying the trait for flower color, Mendel observed that the flowers of each pea plant were either purple or white—but never an intermediate between the two colors. These different, discrete versions of the same gene are called alleles.

http://en.wikipedia.org/wiki/Genetics

Page 6: Genetic Algorithms

6

EVOLUTIONARY COMPUTING (EC)• Evolutionary computing models the processes of

natural evolution.• It is a computer-based problem solving systems that

use computational models of evolutionary processes, such as natural selection, survival of the fittest and reproduction.

Page 7: Genetic Algorithms

7

EVOLUTIONARY ALGORITHM PARADIGMS

•Search or optimization based on genetic evolution: natural selection. By creating a population of solutions and applying genetic operators such as crossover and mutation to evolve the solutions in order to find the best one.

Genetic algorithms

•Evolution of a large number of randomly created computer programs to create program that solves a high-level problem.

•Based on genetic algorithms where individuals are programs represented as trees.

Genetic programming

•Derived from the simulation of adaptive behavior in evolution (i.e., phenotypic evolution).•Mostly applied to real-valued representations.

Evolutionary programming•Modeling the strategic parameters that control variation in evolution, i.e., the evolution of evolution.•For real-valued representation.

Evolutionary strategies

•Similar to genetic algorithms, differing in the reproduction mechanism used. Used for optimization of multi-dimensional real-valued functions.

Differential evolution

•Models the evolution of culture of a population and how the culture influences the genetic and phenotypic evolution of individuals.

Cultural algorithms

•Initially “dumb” individuals evolve through cooperation, or in competition with one another, acquiring the necessary characteristics to survive.

Coevolution

Page 8: Genetic Algorithms

8

GENETIC ALGORITHMS (GA)

Page 9: Genetic Algorithms

9

INTRODUCTION TO GA• Genetic algorithms imitate natural optimization

process, natural selection in evolution.• Developed by John Holland at the University of

Michigan for machine learning in 1975.• Mostly for binary representations.

Page 10: Genetic Algorithms

10

EVOLUTIONARY SEARCH PROCESS

Initiation: Selection of

initial population of chromosomes

Evaluation of the fitness of

each chromosome

Checking for stopping criteria

Selection of chromosomes

Applying genetic

operators

Creating a new

population

Presentation of the best

chromosome

Page 11: Genetic Algorithms

11

Start

Initiation: Selection of initial population of chromosomes

Evaluation of the fitness of chromosomes in the population

Stopping criterion

Selection of chromosomes

Application of genetic operators

Creating a new population

Stop

Presentation of the “best” chromosome

No Yes

Page 12: Genetic Algorithms

12

SELECTION (ROULETTE WHEEL)• The fittest individuals must

have the greatest chance of survival.

• Probability of being selected

http://www.edc.ncl.ac.uk/highlight/rhjanuary2007g02.php/

N

jj

ii

f

fp

1

Page 13: Genetic Algorithms

13

GENETIC OPERATORS• Crossover: combination of genetic material

randomly selected from two or more parents.

• Mutation: process of randomly changing the values of genes in a chromosome.