Ppt on Genetic

Embed Size (px)

Citation preview

  • 8/3/2019 Ppt on Genetic

    1/17

    INTRODUCTION TO GENETIC

    ALGORITHMS

    GROUP

    RAKESH CHAORSIA-090101134(1-7)

    SHUBHAM LOHAN-090101166(8-11)

    RAVIKANT BIHARI-090101136(12-

    16)

  • 8/3/2019 Ppt on Genetic

    2/17

    GA CONCEPT

    Genetic algorithm (GA) introduces theprinciple of evolution and genetics into searchamong possible solutions to given problem.

    This is done by the creation within a machineof a population of individuals represented bychromosomes,in essence a set of character strings, that areanalogous to the DNA, that we have in our

    own chromosomes.

  • 8/3/2019 Ppt on Genetic

    3/17

    BASIC GENETIC ALGORITHM

    Start with a large population of randomly generated

    attempted solutions to a problem

    Repeatedly do the following:

    Evaluate each of the attempted solutions

    Keep a subset of these solutions (the best ones) Use these solutions to generate a new population

    Quit when you have a satisfactory solution (or you run out of

    time)

  • 8/3/2019 Ppt on Genetic

    4/17

    ALGORITHMIC PHASES

    Evaluate Fitness

    Yes

    Initializepopulation

    SatisfyConstrai

    ns

    Select SurvivorsModify Individuals

    Evaluate Fitness

    No

  • 8/3/2019 Ppt on Genetic

    5/17

    CROSSOVER

    Crossover is the similar to naturalreproduction.

    Crossover combines genetic material from twoparents,in order to produce superior offspring.

    Few types of crossover:

    One-point

    Multiple point.

  • 8/3/2019 Ppt on Genetic

    6/17

    CROSSOVER

    E.g.

    6 / 76

    ParentParent 11 Parent 2Parent 2

    00

    11

    5533

    55

    44

    77

    66

    77

    66

    2244

    22

    33

    00

    11

  • 8/3/2019 Ppt on Genetic

    7/17

    CROSSOVER

    E.g.

    7 / 76

    00

    11

    2233

    55

    44

    77

    66

    77

    66

    5544

    22

    33

    00

    11

  • 8/3/2019 Ppt on Genetic

    8/17

    MUTATION

    Mutation introduces randomness into thepopulation.

    Why MutationThe idea of mutation is to reintroduce divergence

    into a converging population.

    Mutation is performed on small part ofpopulation, in order to avoid entering unstablestate.

    In order to ensure that the individuals are notall exactly the same, you allow for a smallchance of mutation.

  • 8/3/2019 Ppt on Genetic

    9/17

    MUTATION

    11 11 00 11 00 1100 00

    00 11 00 11 00 1100 11

    11 00

    00 11

    ParentParent

    ChildChild

  • 8/3/2019 Ppt on Genetic

    10/17

    FITNESS FUNCTION

    Fitness Function is the evaluation function thatis used to evaluated the solutions and find outthe better solutions.

    Fitness of computed for each individual basedon the fitness function and then determinewhat solutions are better than others.

    The fitness function is always problemdependent.

  • 8/3/2019 Ppt on Genetic

    11/17

  • 8/3/2019 Ppt on Genetic

    12/17

    SELECTION

    The selection operation copies a singleindividual, probabilistically selected based onfitness, into the next generation of thepopulation.

    Certain selection methods rate the fitness ofeach solution and preferentially select thebest solutions. Other methods rate only arandom sample of the population, as this

    process may be very time-consuming. Several possible ways Keep the strongest

    Keep some of the weaker solutions

  • 8/3/2019 Ppt on Genetic

    13/17

    SELECTIONSURVIVAL OF THE STRONGEST

    13 / 76

    0.930.93 0.510.51 0.720.72 0.310.31 0.120.12 0.640.64

    Previous generationPrevious generation

    Next generationNext generation

    0.930.93 0.720.72 0.640.64

  • 8/3/2019 Ppt on Genetic

    14/17

    STOPPING CRITERIA

    Final problem is to decide when to stopexecution of algorithm.

    Two possible ways.

    First approach:Stop after production of definite number of

    generations

    Second approach:

    Stop when the improvement in average fitnessover two generations is below a threshold

  • 8/3/2019 Ppt on Genetic

    15/17

    ADVANTAGES OF GA

    Advantages:-

    It can solve every optimization problem whichcan be described with the chromosomeencoding.

    It solves problems with multiple solutions.

    Genetic algorithms are easily transferred toexisting simulations and models.

    Genetic algorithm is a method which is veryeasy to understand and it practically does notdemand the knowledge of mathematics.

  • 8/3/2019 Ppt on Genetic

    16/17

    DISADVANTAGES OF GA

    Certain optimization problems (they are calledvariant problems) cannot be solved by meansof genetic algorithms. This occurs due topoorly known fitness functions which generate

    bad chromosome blocks in spite of the factthat only good chromosome blocks cross-over.

    There is no absolute assurance that a geneticalgorithm will find a global optimum. It

    happens very often when the populationshave a lot of subjects.

  • 8/3/2019 Ppt on Genetic

    17/17

    THANK YOU!!!