Evolutionary Computation and Artificial Life - Grader comments for exercise 1

Report structure

• Evolutionary parameters
  When you write your reports, keep in mind it should enable the reader to understand what you did without exploring your code. In particular, make sure you state the following:
  • Population size
  • Generation count or termination conditions
  • Genome representation
  • Fitness calculation function
  • Crossover method and crossover rate
  • Mutation method and mutation rate
  • Selection method
  • any other evolutionary parameter you have used...
  In this assignment some parameters (such as XO & Mutation) were already defined, so there's no need to restate them in your reports.
• Graphs
  Graphs should be included within the report body, with attached explanations. (To add graphs to html documents, save your graphs as image files, then link them to your reports). Some notes on graphs:
  • Make sure to include axes captions, legend and title.
  • If your report is in English, the x-axis should ascend from left to right. The Hebrew Excel version tends to reverse the x-axis, but you can fix this by double-clicking the x-axis, and unchecking the 'categories in reverse order' box (under the 'scale' tab).
  • When you list several graphs side by side, ensure their scale is the same. In other words, the min & max values of the axes should be identical, so the viewer can see the difference between the two graphs.
• Results, Discussion & Conclusions
  You should discuss the results you got from your algorithm. The discussion includes the actual results (for example: best knapsack value in this exercise), as well as other findings and observations (for example: which XO & Mutation rates gave out the best results).
  You should try to explain why, in your opinion, you got these results. Don't be afraid to give your explanations, in most cases there is no obvious right and wrong answers - as long as your explanations are logical and rely on actual results, the will be acceptable.
  An important note: Some of you got bad results due to a bug. Though this course doesn't focus on programming, you should be able to tell when your results are invalid and debug your code. This is a serious fault, as your entire work relies on these results! (For example, if your fitness decreases during time, or give out random values regardless of the evolutionary parameters you try, there must be a bug somewhere within your code...)

Submission & Presentation

• The entire report should be readable via standard web browser. For this sake, post your works in one of the following formats:
  • html (best)
  • pdf
  Please avoid posting Microsoft Word's doc files, as they are not readable on unix systems. You can edit your work with Word, then save it as html. You can also convert doc files to pdf using free online tools such as PDFOnline.
• The submitted URL should point to the exercise solution. Please don't point to a directory that contains the work files.
• Don't zip your works. Your code may be zipped, if you wish.
• When you post your code, include only source files (+ data files, readme, etc... as you wish) - avoid including class files and compiled code.