ECAL-081 - Assignment 5 Grades


The table shows the grades for Assignment 5.

Detailed notes can be found at the bottom of the page. To see them:
- enable ActiveX controls (yellow bar at the top of the page, in IE)
- click on the grade to toggle my comments on and off.
      Total
ID Grade Bonus Points
066197781 Good   90
043397561 Good   90
040600314 Excellent 2 102
034077636 Excellent 3 103
043557214 Excellent 6 106
043246495 Excellent 6 106
061997292 Very Good   95
309190585 Very Good   95
060643665 Very Good_Minus   92
053084489 Very Good_Minus   92
040144594 Excellent   100
036155679 Excellent   100
061109013 Excellent   100
040296535 Excellent   100
040680993 Excellent   100
060403144 Excellent 2 102
060414562 Excellent 2 102
011950466 Excellent 3 103
040210148 Excellent 3 103
066096520 Very Good_Plus   97
040952061 Very Good_Plus   97
304606536 Very Good   95
312817927 Very Good   95
037770039 Very Good_Plus   97
304369283 Excellent   100



AuthorsQGrade
066197781,043397561Q1Good
  • Nice report, ordered and readable
  • I cannot understand why you got poor results. Your 2nd fitness function looks pretty good. Maybe there is a bug in your code? Too bad you did not have time to fix it
  • Interesting conclusions
040600314Q1Excellent
  • Very interesting report
  • Very interesting discussion and process of work
  • Interesting conclusions
  • Nice trick with the first and last rules
  • +2 points bonus on excellent report
034077636Q1Excellent
  • Very interesting report
  • Discussion and conclusions capture your work nicely
  • Nice diagrams (how do you do that in Excel?) and graphs
  • +3 points bonus on excellent report
043557214,043246495Q1Excellent
  • Very good report
  • Like the monkey picture :-)
  • Very nice extra work, and good job changing to a harder problem, Density, to show the improvements in action.
  • "Non Symmetrical Connections" does seem like a natural extension, great to see you get better reaults.
  • "Double moral" is also a nice extension, but you do not explain at what times the Boolean is flipped. From looking at your I understand that it is flipped every 3 "generations", and wonder if I would have expected the improvement in results. Anyway, good job!
  • Interesting conclusions
  • +6 points bonus for extra work and great reprot
061997292,309190585Q1Very Good
  • Nice report
  • Good parameters description
  • Missing a but of discussion and conclusions
060643665,053084489Q1Very Good_Minus
  • Overall good report, but poor results
  • The syncronization problem should not be too hard to solve. Maybe you had a bug in your code, or needed some more experimenting with the fitness function
  • Missing some more conclusions
040144594Q1Excellent
  • Very interesting report
  • Nice idea of distance evolution. I wonder if the results would have been different without it.
  • Nice conclusions
036155679,061109013Q1Excellent
  • Original fitness funciton, though the convergence to a good CA seems relatively slower
  • Nice touch with graphically showing how evolution grew better CA's
  • Altogether very interesting report
040296535,040680993Q1Excellent
  • Nice detailed report
  • Interesting discussion and conclusions
060403144,060414562Q1Excellent
  • Very nice, ordered, readable report
  • Very interesting discussion and process of work
  • Interesting conclusions
  • Nice rug
  • +2 point bonus on great report
011950466,040210148Q1Excellent
  • Very nice, readable report
  • Very interesting discussion and process of work
  • Very interesting conclusions
  • Mutation in each configuration, means mutating at a low probability each configuration. Glad you found that out empirically
  • Great you showed and analyzed also the rule-code distribution between the cells
  • +3 points bonus on excellent report
066096520,040952061Q1Very Good_Plus
  • Interesting, good report
  • Nice touch with graphically showing how evolution grew better CA's
  • A bit confusing to use the term "generation", once for 300 and conce for 1500 configurations
  • Good conclusions
304606536,312817927Q1Very Good
  • Interesting, good report
  • Nice to see the "evolution" of fitness function in your work
  • Missing some discussion and conclusions
  • Small note: the evolution on the rules is done every K iterations, not every C mod K.
037770039Q1Very Good_Plus
  • Interesting, good report
  • Nice 3D graphs, missing some explanations - what are S1, S2, "Generations"?
  • Missing some discussion and conclusions
304369283Q1Excellent
  • Very interesting report
  • Nice graphs and histograms
  • Interesting conclusions
  • Great you showed and analyzed also the rule-code distribution between the cells