Warm-up Activity 1. How many frames are in a Pixar animated movie such as The Incredibles?
Genetic Algorithms: “Natural Selection”
Genetic Algorithms HISTORY: 1960sEvolutionary computing used to solve complex engineering problems by Rechlenberg 1970sGenetic algorithms invented by John Holland 1980sGE begins selling first genetic algorithm product 1992John Koza invents genetic programming
Genetic Algorithms Genetic algorithms have lots of real world applications: Automotive car design for composite materials and aerodynamics simultaneously
Genetic Algorithms Genetic algorithms have lots of real world applications: Engineering design of complex components, structures and operations (e.g. heat exchanger optimization, turbines, building trusses).
Genetic Algorithms Genetic algorithms have lots of real world applications: Evolvable Hardware - electronic circuits created by GA computer models that use stochastic (statistically random) operators to evolve new configurations from old ones.
Genetic Algorithms Genetic algorithms have lots of real world applications: Encryption and Code Breaking- GAs can be used both to create encryption for sensitive data as well as to break those codes
Genetic Algorithms Genetic algorithms have lots of real world applications: Molecular Design - GA optimization and analysis is used for designing industrial chemicals or for proteins used in pharmaceuticals.
Genetic Algorithms Genetic algorithms have lots of real world applications: Biomimetics - GA optimization and analysis is used in the development of technologies inspired by designs in nature.
Genetic Algorithms Genetic algorithms have lots of real world applications: Linguistics- GA can be used to generate puns or even help write jokes!
Genetic Algorithms STRENGTHS: Good at finding solutions quickly Capable of finding multiple solutions Can solve problems that are not well understood
Genetic Algorithms WEAKNESSES: Doesn’t discriminate between local and global minimums No guarantee of finding the best solution; only returns “good” soluton Difficult to predict performance; requires a lot of fine tuning
Genetic Algorithms Genetic algorithms usually consist of the following five steps: 1.Create a starting population randomly 2.Test the fitness of each member and assign selection probability 3.Reproduce 4.Test new population for threshold criteria 5.Wash, rinse and repeat…
Genetic Algorithms Reproduction: – Select two parent chromosomes from a population according to their fitness) – Cross over the parents to form a new offspring (children). – Mutate new offspring at each locus (position in chromosome). – Place new offspring in a new population
Genetic Algorithms Now let’s put this to work… X 3 – Y 2 + Z = 25 Let’s find a solution set [X,Y,Z] for this equation as a class by using a simple GA routine. You’ll need a pencil and maybe a calculator.