{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

Genetic Algorithms

# Genetic Algorithms - ERE284 GeneticAlgorithms...

This preview shows pages 1–10. Sign up to view the full content.

Genetic Algorithms Genetic Algorithms Applications to Reservoir Engineering Applications to Reservoir Engineering ERE 284 ERE 284 Barı ş  G ű yag ű ler Burak Yeten

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
G ü yag ü ler Genetic Algorithm Aspects Developed by John Holland, 1975 - SGA Many developments since then Applications to real world problems Hybridization with other algorithms  Mimics mechanics of natural selection and  natural genetics
G ü yag ü ler Genetic Algorithm (GA) Create initial population N Return best Y Select individuals Apply GA operators: crossover, mutation Stop  criteria  met?

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
G ü yag ü ler GA Vocabulary Individual    011...0010     x     Fitness = f(x ) Chromosome Problem parameters Evaluation Function Generation 1 Generation 2 Generation  N Individual 1 Individual 2 Individual  n Individual 1 Individual 2 Individual  n Individual 1 Individual 2 Individual  n
G ü yag ü ler GA Data Structures Binary Variables decoded into  0 ’s and  1 ’s Arranged linearly 0011010011000111 x 1 x 2 x 3 x 4 Integer/Float Combination

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
G ü yag ü ler Operators Crossover  010111 ,100101 One point  0101 11 ,1001 01 0101 01,1001 11 Two point 0 101 11 ,1 001 01 0 001 11 ,1 101 01 Uniform 010111 ,100101 1 10 1 1 1, 0 00 1 0 1 Mutation Bit inversion 1 0 0 1 Bit reinitialization  1,0 ⇒?
G ü yag ü ler GAs, Why Do They Work? Some terms: Schema introduced by ‘ * ’ (wildcard) 0**10 - 0 00 10,0 01 10,0 10 10,0 11 10 Order of a schema, o( binary string ) o( 0**10 ) = 3 Length of a schema,  δ ( binary string ) δ ( 0**10 ) = 5-1 = 4

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
G ü yag ü ler GAs, Why Do They Work? The Schema Theorem:  Selection based  on  fitness  increases the sampling rate of  above-average building-blocks  exponentially Crossover  introduces new building-blocks  and enables information exchange Mutation  introduces new building-blocks
G

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}