Genetic Algorithms
Download as PPTX, PDF12 likes13,437 views
This document discusses genetic algorithms and their applications. It explains key concepts like genetic crossover, genetic algorithm steps to solve optimization problems, and how genetic algorithms mimic biological evolution. Examples are provided of genetic algorithms being used for tasks like predicting protein structure, automotive design optimization, and generating musical variations. Advantages and limitations of genetic algorithms are also summarized.
Genetic Algorithms
- 1. By Deepali Kundnani Shruti Railkar
- 2. Survival of the Fittest Natural selection Sir Charles Darwin
- 3. Chromosomes from two different parents Chromatids from each overlap at Chiasma Recombinant chromosomes are form Further passed on to the progeny Genetic Crossover
- 4. A T T G C T C ORIGINAL A T A G C T C SUBSTITUTION A T T G A C T C ADDITION A T G C T C DELETION
- 5. Offsprings have combinations of features inherited from each parent Image adapted from http://www.wpdipart.com
- 6. Random changes are observed Image adapted from http://www.wpdipart.com
- 7. Genetic Algorithm is a type of local search that mimics evolution by taking a population of strings which encode possible solutions and combines them based on a fitness function to produce individuals that are more fit.
- 8. 1) Encoding the two numbers into binary strings Parent 1=3.273672 =>11.0100011000001 Parent 2=3.173294 =>11.0010110001011 2) Randomly choose a crossover point; let suppose be it at bit 6, and we split the gene at position six. Parent 1=>3.273672=>11.010---0011000001 Parent 2=>3.173294=>11.001---0110001011 3) Swapping the two tails ends of binary strings. Child 1=>11.010---0110001011 Child 2=>11.001---0011000001 4) Recombining the two binary strings to get two new offspring. Child 1=>11.0100110001011 Child 2=>11.0010011000001 5) Decoding the binary strings back into floating point numbers. Child 1=3.298218 Child 2=3.148560
- 11. First Generation
- 12. 60th Generation
- 13. 95th Generation
- 14. 100th Generation
- 15. Artificial Intelligence Automotive Design Computer Gaming Predicting Protein Structure Optimization Problems Music Business
- 16. Helps to determine the accurate torsion angles and predict protein structure
- 17. Evolution of Monalisa : Roger Alsig Weblog
- 18. Minimizing total error over the set of data points Source:http://www.geneticprogramming.org
- 19. Musical examples of variations output to get perfect music. Fitness function determinant here is human ear
- 20. Optimization of aerodynamics of a Car for a smooth drive on a crooked path Source: Youtube
- 25. ADVANTAGES LIMITATIONS No training required Do not work well when the population size is small and the rate of change is too high. Efficient even during Multi-modal or n-dimensional search space If the fitness function is chosen poorly or defined vaguely, the Can work for non-linear Genetic Algorithm may be equations too unable to find a solution to the problem, or may end up Efficient solving the wrong problem
- 26. GAOT- Genetic Algorithm Optimization Toolbox in Matlab JGAP is a Genetic Algorithms and Genetic Programming component provided as a Java framework Generator is another popular and powerful software running on Microsoft Excel
- 27. Genetic Algorithm is related to “solving problems of everyday interest” in many diverse fields. However, several improvements can be made in order that Genetic Algorithm could be more generally applicable. Future work will continue through evolution and many more specific tasks
- 28. Introduction to Genetic Algorithms -Axcelis http://www.axcelis.com:80/articles/itga/application.html How Genetic Algorithm works http://www.mathworks.in/help/toolbox/gads/f6187.html Introduction to Bioinformatics By Sundararajan & R. Balaji Functioning of a Genetic Algorithm http://www.rennard.org/alife/english/gavintrgb.html#gafunct