genetic algorithms k.ganesh reasearch scholar, ph.d., industrial management division, humanities and...
TRANSCRIPT
Genetic Algorithms
K.GaneshK.Ganesh
Reasearch Scholar, Ph.D.,Reasearch Scholar, Ph.D.,
Industrial Management Division,Industrial Management Division,
Humanities and Social Sciences Humanities and Social Sciences Department,Department,
Indian Institute of Technology Madras,Indian Institute of Technology Madras,
Chennai,TN,India.Chennai,TN,India.
Overview
IntroductionIntroduction HistoryHistory DetailsDetails ExampleExample FutureFuture
Introduction• Optimization has for long been an important goal in the fields of Applied Mathematics and Computer Science.
• As the name suggests this method is based on Darwin’s Theory of evolution.
•Genetic algorithms arose from computer simulations of biological evolution in the late 60s and early 70s.
•Genetic algorithms are a part of evolutionary computing, which is a rapidly growing area of artificial intelligence
History
• 1960 – Introduced by 1960 – Introduced by I. Rechenberg I. Rechenberg
• 1975 – Popularized by John Holland1975 – Popularized by John Holland
• 1975 - book "Adaptation in Natural and Artificial 1975 - book "Adaptation in Natural and Artificial Systems" published Systems" published
• 1992 – John Koza’s work1992 – John Koza’s work
Details
A description of the biological terms used:A description of the biological terms used:
1.1. ChromosomesChromosomes
1.1. GenesGenes
2.2. LocusLocus
2.2. ReproductionReproduction
1.1. CrossoverCrossover
2.2. MutationMutation
1.1. StartStart2.2. Fitness Fitness 3.3. New PopulationNew Population
1.1. SelectionSelection2.2. CrossoverCrossover3.3. MutationMutation
4.4. AcceptingAccepting4.4. ReplaceReplace5.5. TestTest6.6. LoopLoop
•General Outline
1. Encoding
Chromosome 1 1101100100110110
Chromosome 2 1101111000011110
2. Fitness & Selection
• The fitness function f(x)
• Associates fitness of a chromosome to a single number
• This number determines the chance of selection for reproduction
4. Mutation
Original offspring 1
1101111000011110Original offspring 2
1101100100110110Mutated offspring 1
1100111000011110Mutated offspring 2
1101101100110110
Chromosome 1 11011 | 00100110110
Chromosome 2 11011 | 11000011110
Offspring 1 11011 | 11000011110
Offspring 2 11011 | 00100110110
3. Crossover
Example
• Character Evolution AlgorithmCharacter Evolution Algorithm1.1. Generate some random individualsGenerate some random individuals2.2. Select the n best of them depending on their fitnessSelect the n best of them depending on their fitness 3.3. Take those n best to produce some new individuals, based on the Take those n best to produce some new individuals, based on the
information they hold. information they hold. Repeat Repeat from step 2, until you reach what you want.from step 2, until you reach what you want.
• FitnessFitness
• Reproduction & Cumulative SelectionReproduction & Cumulative Selection
• MutationMutation
• View AppletView Applet
Applications
• Decision MakingDecision Making• Data MiningData Mining• SchedulingScheduling• Computer gamesComputer games• Stock Market TradingStock Market Trading• MedicalMedical• Information Systems ApplicationsInformation Systems Applications • Finance ApplicationsFinance Applications
References
• Cawsey, Allison. The essence of A.I.. Prentice Hall. 1998Cawsey, Allison. The essence of A.I.. Prentice Hall. 1998• Introduction to Genetic Algorithms.Introduction to Genetic Algorithms.
http://cs.felk.cvut.cz/~xobitko/ga/http://cs.felk.cvut.cz/~xobitko/ga/• Applications of Genetic Algorithms. Applications of Genetic Algorithms.
http://www.doc.ic.ac.uk/~nd/surprise_96/journal/vol1/tcw2/article1.htmlhttp://www.doc.ic.ac.uk/~nd/surprise_96/journal/vol1/tcw2/article1.html• Genetic AlgorithmsGenetic Algorithms
http://http1.brunel.ac.uk:8080/depts/AI/alife/ga.htmhttp://http1.brunel.ac.uk:8080/depts/AI/alife/ga.htm• Evolutionary Algorithms Evolutionary Algorithms
http://www2.informatik.uni-erlangen.de/IMMD-II/Persons/jacob/Evolvica/http://www2.informatik.uni-erlangen.de/IMMD-II/Persons/jacob/Evolvica/Java/CharacterEvolution/index.htmlJava/CharacterEvolution/index.html
• Genetic Algorithm, Ashish Gupta.Genetic Algorithm, Ashish Gupta.• Genetic Algorithm and Classifier System, David Goldstein. Genetic Algorithm and Classifier System, David Goldstein.