Scribd Logo
Upload

Welcome to Scribd!

  • 49 views

    Schema Theory: Suggested Reading: D. E. Goldberg, Genetic Algorithm in

    Uploaded by

    Praveen Kumar
    This document summarizes key concepts from a lecture on schema theory in genetic algorithms. It defines schemas as templates that represent sets of binary strings, and discusses their order, defining length, and how they are counted. It then analyzes the effects of genetic operators like selection, crossover and mutation on schemas, and presents the schema theorem, which states that schemas with above average fitness, short defining lengths and lower order are more likely to survive from one generation to the next.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Schema Theory: Suggested Reading: D. E. Goldberg, Genetic Algorithm in

    Uploaded by

    Praveen Kumar
    49 views21 pages
    This document summarizes key concepts from a lecture on schema theory in genetic algorithms. It defines schemas as templates that represent sets of binary strings, and discusses their order, defining length, and how they are counted. It then analyzes the effects of genetic operators like selection, crossover and mutation on schemas, and presents the schema theorem, which states that schemas with above average fitness, short defining lengths and lower order are more likely to survive from one generation to the next.

    Original Description:

    schema

    Original Title

    schema

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document summarizes key concepts from a lecture on schema theory in genetic algorithms. It defines schemas as templates that represent sets of binary strings, and discusses their order, defining length, and how they are counted. It then analyzes the effects of genetic operators like selection, crossover and mutation on schemas, and presents the schema theorem, which states that schemas with above average fitness, short defining lengths and lower order are more likely to survive from one generation to the next.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    49 views21 pages

    Schema Theory: Suggested Reading: D. E. Goldberg, Genetic Algorithm in

    Uploaded by

    Praveen Kumar
    This document summarizes key concepts from a lecture on schema theory in genetic algorithms. It defines schemas as templates that represent sets of binary strings, and discusses their order, defining length, and how they are counted. It then analyzes the effects of genetic operators like selection, crossover and mutation on schemas, and presents the schema theorem, which states that schemas with above average fitness, short defining lengths and lower order are more likely to survive from one generation to the next.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 21

    Lecture 3:

    Schema Theory
    Suggested reading: D. E. Goldberg, Genetic Algorithm in
    Search, Optimization, and Machine
    Learning, Addison Wesley Publishing
    Company, January 1989

    Schema Theorem

    Schema theorem serves as the analysis tool for


    the GA process

    Explain why GAs work by showing the


    expectation of schema survival
    Applicable to a canonical GA
    binary representation
    fixed length individuals
    fitness proportional selection
    single point crossover
    gene-wise mutation

    Schema
    A schema is a set of binary strings that match
    the template for schema H
    A template is made up of 1s, 0s, and *s where
    * is the dont care symbol that matches either
    0 or 1

    Schema Examples

    The schema H = 10*1* represents the set of


    binary strings
    10010, 10011, 10110, 10111

    The string 10 of length l = 2 belongs to 2l =


    22 different schemas
    **, *0, 1*, 10
    4

    Schema: Order o(H)

    The order of a schema is the number of its


    fixed bits, i.e. the number of bits that are not
    * in the schema H

    Example: if H = 10*1* then o(H) = 3

    Schema: Defining Length (H)

    The defining length is the distance between its


    first and the last fixed bits

    Example: if H = *1*01 then (H) = 5 2 = 3

    Example: if H = 0**** then (H) = 1 1 = 0

    Schema: Count

    Suppose x is an individual that belongs to the


    schema H, then we say that x is an instance of
    H (xH)

    m(H, k) denotes the number of instances of H in


    the k th generation

    Schema: Fitness

    f(x) denotes fitness value of x

    f(H,k) denotes average fitness of H in the k-th


    generation
    f (H , k ) =

    f ( x)

    xH

    m( H , k )
    8

    Effect of GA On A Schema
    Effect of Selection
    Effect of Crossover
    Effect of Mutation
    = Schema Theorem

    Effect of Selection on Schema

    Assumption: fitness proportional selection

    Selection probability for the individual x


    ps ( x) =

    f ( x)
    N

    f (x )
    i =1

    where the N is the total number of individuals


    10

    Net Effect of Selection

    The expected number of instances of H in the


    mating pool M(H,k) is
    f (H , k )

    xH f ( x )
    M (H , k ) =
    = m( H , k )
    f

    Schemas with fitness greater than the


    population average are likely to appear
    more in the next generation
    11

    Effect of Crossover on Schema

    Assumption: single-point crossover

    Schema H survives crossover operation if


    one of the parents is an instance of the
    schema H AND
    one of the offspring is an instance of the
    schema H

    12

    Crossover Survival Examples


    Consider H = *10**
    P1 = 1 1 0 1 0 H

    S1 = 1 1 0 1 1 H Schema H

    P2 = 1 0 1 1 1 H

    S2 = 1 0 1 1 0 H

    P1 = 1 1 0 1 0 H

    S1 = 1 1 1 1 1 H Schema H

    P2 = 1 0 1 1 1 H

    S2 = 1 0 0 1 0 H

    survived

    destroyed

    13

    Crossover Operation

    Suppose a parent is an instance of a schema H.


    When the crossover is occurred within the bits
    of the defining length, it is destroyed unless the
    other parent repairs the destroyed portion

    Given a string with length l and a schema H


    with the defining length (H), the probability
    that the crossover occurs within the bits of the
    defining length is (H)/(l - 1)

    14

    Crossover Probability Example


    Suppose

    H = *1**0

    We

    gave
    l=5
    (H) = 5 2 = 3
    Thus, the probability that the crossover
    occurs within the defining length is 3/4

    15

    Crossover Operation

    The upper bound of the probability that the


    schema H being destroyed is
    Dc ( H ) pc

    (H )
    l 1

    where pc is the crossover probability

    16

    Net Effect of Crossover

    The lower bound on the probability Sc(H) that


    H survives is

    S c ( H ) = 1 Dc ( H ) 1 pc

    (H )
    l 1

    Schemas with low order are more


    likely to survive
    17

    Mutation Operation

    Assumption: mutation is applied gene by gene

    For a schema H to survive, all fixed bits must


    remain unchanged

    Probability of a gene not being changed is

    (1 pm )
    where pm is the mutation probability of a gene
    18

    Net Effect of Mutation

    The probability a schema H survives under


    mutation

    S m ( H ) = (1 pm ) o ( H )
    Schemas with low order are more
    likely to survive

    19

    Schema Theorem
    Exp. # of Schema H in Next Generation >
    f (H , k )
    Exp. # in Mating Pool ( M ( H , k ) = m( H , k )
    )
    f
    (H )

    Prob. of Surviving Crossover ( S c ( H ) 1 pc

    l 1

    Prob. of Surviving Mutation ( S m ( H ) = (1 pm ) o ( H ) )

    20

    Schema Theorem

    Mathematically

    f (H , k )
    (H )
    o( H )
    E[ m ( H , k + 1)] m ( H , k )
    1
    p
    (
    1

    p
    )

    c
    m
    f
    l 1

    The schema theorem states that the


    schema with above average fitness, short
    defining length and lower order is more
    likely to survive
    21

    You might also like