Due: Friday, May 4th

• Finish the final research paper for the course.
• Pick a question to address using your algorithm. The simplest types of questions depend on a parameter of the GA. For example, which of two different types of mutation give the best results. Or two different forms of crossover or selection. You should frame this as a hypothesis, for example: "Our research hypothesis is that using a tournament of size 5 will lead to better fitness values than a tournament size of 3."
• Using the IEEE template from Assignment 9 create a research paper outline. Include sections for the abstract, introduction, methods, results, and conclusion (no background).
• Revise the introduction for your paper. Very generally what is a genetic algorithm. What is the importance of the component of the genetic algorithm that you are experimenting with? For example, if your experiment focuses on the selection set discuss the importance of selection. Then present the hypothesis you are testing.
• Revise the methods for your paper. Describe your algorithms, including how selection, crossover, and mutation are done. Describe your experiments. Include a table with the parameters of your algorithm: population size, number of generations, mutation rate (probability of a value mutating), etc.
• Add an abstract to your paper. This should briefly describe the motivation for your paper, the experiment that you did, and the results.
• Add a results section to your paper. At a mininimum this should include a graph showing average fitness over time for each of your test cases and a table showing the average fitness and standard deviation of the fitness in the last generation for each of your test cases. The text in the results section should refer to both the graph and the table and describe what the read should be observing.
• Add a conclusion section to your paper. The conclusion should summarize what you did and what the results were.
Turn in: The pdf of your research paper and the Python code used to generate the data.