Optimizing Genetic Algorithms for Time Critical Problems

Genetic algorithms have a lot of properties that makes it a good choice when one needs to solve very complicated problems. The performance of genetic algorithms is affected by the parameters that are used. Optimization of the parameters for the genetic algorithm is one of the most popular research fields of genetic algorithms. One of the reasons for this is because of the complicated relation between the parameters and factors such as the complexity of the problem. This thesis describes what happens when time constraints are added to this problem.

1 Introduction
1.1 History
1.2 Genetic Algorithms
1.2.1 The genetic information
1.2.2 Fitness function
1.2.3 Selection
1.2.4 Reproduction
1.2.5 Crossover
1.2.6 Mutation
1.2.7 Other operations
1.3 Where are genetic algorithms used?
1.4 Optimization of the population size
1.4.1 Optimization based on generations
1.4.2 Optimization based on time constraint
1.5 Hypothesis
1.6 Methodology
1.7 Outline of the thesis
2 Problem domain
2.1 Controlling and Designing a time critical system
2.2 Genetic algorithms to solve time critical problems
3 Genetic Algorithm Parameters
3.1 Exploitation vs. exploration
3.1.1 Mutation rate
3.1.2 Crossover rate
3.1.3 Adaptive genetic algorithms
3.2 Population size
4 Experimental setup
4.1 Choice of genetic algorithm
4.1.1 Selection
4.1.2 Mutation and Crossover
4.1.3 Parameter to binary representation
4.1.4 Random
4.2 De Jong test functions
i4.2.1 Test function 1 or Sphere
4.2.2 Test function 2 or Rosenbrock’s Saddle
4.2.3 Test function 3 or Step
4.2.4 Test function 4 or Quartic
4.2.5 Test function 5 or Shekel’s Foxholes
4.3 Execution of the test
4.3.1 Log file generation
4.3.2 Graph transformation
4.3.3 Maximum fitness generation
5 Analysis of the results
5.1 Explanation of the figures
5.2 Analysis of the results
5.3 Time and Performance measuring
5.4 Comparison to others studies
5.5 Function to find optimal population size
5.6 De Jong 5
5.7 Functions characteristics
6 Discussion
7 Conclusions and future work
7.1 Future Work

Author: Christian Johansson, Gustav Evertsson

Source: Blekinge Institute of Technology

Reference URL 1: Visit Now

1 thought on “Optimizing Genetic Algorithms for Time Critical Problems”

Leave a Comment