11
Exploring Genetic Algorithms Through the Iterative Prisoner's Dilemma Computer Systems Lab 2007-2008 Aaron Dufour

Exploring Genetic Algorithms Through the Iterative Prisoner's Dilemma Computer Systems Lab 2007-2008 Aaron Dufour

Embed Size (px)

Citation preview

Page 1: Exploring Genetic Algorithms Through the Iterative Prisoner's Dilemma Computer Systems Lab 2007-2008 Aaron Dufour

Exploring Genetic Algorithms Through the Iterative Prisoner's

Dilemma

Computer Systems Lab 2007-2008Aaron Dufour

Page 2: Exploring Genetic Algorithms Through the Iterative Prisoner's Dilemma Computer Systems Lab 2007-2008 Aaron Dufour
Page 3: Exploring Genetic Algorithms Through the Iterative Prisoner's Dilemma Computer Systems Lab 2007-2008 Aaron Dufour
Page 4: Exploring Genetic Algorithms Through the Iterative Prisoner's Dilemma Computer Systems Lab 2007-2008 Aaron Dufour

Mutation Rate

How the mutation rate changes per generationWithin a generation the mutation rate does not

change

Page 5: Exploring Genetic Algorithms Through the Iterative Prisoner's Dilemma Computer Systems Lab 2007-2008 Aaron Dufour

Natural Selection

How many are removed from the population each generation

Static rate – the same number are removed each generation

Fitness-based rate – all those below a threshold fitness value are removed

Page 6: Exploring Genetic Algorithms Through the Iterative Prisoner's Dilemma Computer Systems Lab 2007-2008 Aaron Dufour

Recombination

SinglePoint1001010101000011Yields1000001101010101

DoublePoint1001010101000011Yields1000000101010111

Page 7: Exploring Genetic Algorithms Through the Iterative Prisoner's Dilemma Computer Systems Lab 2007-2008 Aaron Dufour

Initial Population Creation

SimpleRandom binaryFlip HalfRandom on first halfSecond half is inverted first halfEnsures that every bit has 50% 1's and 50% 0'sCheck for DuplicatesSame as flip half, except remakes each one that

has a duplicateEnsures that all of the solutions are different

Page 8: Exploring Genetic Algorithms Through the Iterative Prisoner's Dilemma Computer Systems Lab 2007-2008 Aaron Dufour

Output

Outputs the average fitness value for each generation File name is “g i p t s m n r f.txt” g – number of generations i – number of iterations p – population size t – number of turns s – initial population type m – mutation rate info n – natural selection info r – recombination type f – test number Example – 10 100 150 s s-0.0050 s-0.5 s t0.txt

Page 9: Exploring Genetic Algorithms Through the Iterative Prisoner's Dilemma Computer Systems Lab 2007-2008 Aaron Dufour

Data Analysis

Page 10: Exploring Genetic Algorithms Through the Iterative Prisoner's Dilemma Computer Systems Lab 2007-2008 Aaron Dufour

Data Analysis, cont’d

The program analyzes the data to find where the fitness stabilizes

Although we can do this visually, it is difficult for the computer

My algorithm eliminates data from the left side until the slope of a fit line gets within a certain amount of 0

Page 11: Exploring Genetic Algorithms Through the Iterative Prisoner's Dilemma Computer Systems Lab 2007-2008 Aaron Dufour

Next Quarter

Next quarter I will automate the process of creating data and then finding the stabilization point

Then I can use the results to come to a conclusion about the different methods