Difference between revisions of "John Hawkins's Term Project Page"

From Computational Statistics (CSE383M and CS395T)
Jump to navigation Jump to search
Line 9: Line 9:
  
 
Comment: <br>
 
Comment: <br>
It is good to see your presentation material. I think it is a good idea to include an example ('Tiddlywinks' part) of the genetic algorithm. I'm not sure if you already have in mind the followings, but I have some suggestions. (1) You can show the graph of 'the best solution performance vs. generations'. It would be helpful to visualize how your solutions are evolved over generations. (2) You can vary your mutation rate, and can see what happens if the mutation rate is high. You can check the results according to the mutation rate, and may argue that it is important to set the reasonably low mutation rate. [[User:Joyce Whang|Joyce Whang]] 02:06, 21 April 2012 (CDT)
+
It is good to see your presentation material. I think it is a good idea to include an example ('Tiddlywinks' part) of the genetic algorithm. I'm not sure if you already have in mind the followings, but I have some suggestions. (1) You can show the graph of 'the best solution performance vs. generations'. It would be helpful to visualize how your solutions are evolved over generations. (2) You can try to vary your mutation rate, and can see what happens if the mutation rate is high. You can check the results according to the mutation rate, and may argue that it is important to set the reasonably low mutation rate. [[User:Joyce Whang|Joyce Whang]] 02:06, 21 April 2012 (CDT)

Revision as of 03:13, 21 April 2012

Video opening: Presentation Opening (I made the audio and compiled online footage. No, unfortunately I didn't make the visuals from scratch. =) )

The current version of my PowerPoint presentation is here: Media:Genetic_Algorithms_Apr9.pptx

The graph in slide 4 comes from a program I have written to illustrate genetic algorithms. My remaining work for the project is to flesh out more details in the PowerPoint presentation, incorporate my example code, and finally, make a webcast.

And just for fun: Media:Opinionated_Lessons_Spread.pdf


Comment:
It is good to see your presentation material. I think it is a good idea to include an example ('Tiddlywinks' part) of the genetic algorithm. I'm not sure if you already have in mind the followings, but I have some suggestions. (1) You can show the graph of 'the best solution performance vs. generations'. It would be helpful to visualize how your solutions are evolved over generations. (2) You can try to vary your mutation rate, and can see what happens if the mutation rate is high. You can check the results according to the mutation rate, and may argue that it is important to set the reasonably low mutation rate. Joyce Whang 02:06, 21 April 2012 (CDT)