KTH / CSC / Courses / DD2445 / DD2445 Autumn 2015 / Course evaluation

Course Evaluation for DD2445 Complexity Theory, Autumn 2015

General questions about the course

How did you follow the course?
1. 89% (8 st) I was taking the course for credit and have done all problem sets.
2. 11% (1 st) I started taking the course for credit but did not finish it.
3. 0% (0 st) I was a listener on the course.
Did you find the course easy or hard?
1. 0% (0 st) Very easy.
2. 0% (0 st) Fairly easy.
3. 11% (1 st) Medium.
4. 56% (5 st) Fairly hard.
5. 33% (3 st) Very hard.
Did you find the course interesting?
1. 67% (6 st) Yes, very interesting.
2. 22% (2 st) Yes, fairly interesting.
3. 11% (1 st) Neutral.
4. 0% (0 st) No, not very interesting.
5. 0% (0 st) No, not at all interesting.
Did you understand sufficiently well at the beginning of the course what the course goals were?
1. 67% (6 st) Yes.
2. 33% (3 st) Don't know.
3. 0% (0 st) No.
If you think this would have been needed, how should the course goals have been explained better?
I found the course goals to be well-defined at the beginning.
In the prerequisites for the course, it said that "you need to have taken DD1352 Algorithms, Data Structures, and Complexity or DD2352 Algorithms and Complexity, or corresponding courses at other universities, and should feel comfortable with that material. There are no additional formal prerequisites on top of what is stated in the Study Handbook, but you will need mathematical maturity and a willingness to learn new stuff." Do you think this was a fair description?
1. 89% (8 st) Yes.
2. 0% (0 st) Don't know.
3. 11% (1 st) No.
What if anything would you have added to or changed in the course prerequisites?
Nothing comes to mind, the course built logically upon the complexity aspects of the course examples, and mathematical maturity was definitely needed.
Do you think you personally had the required background to be able to follow the course?
1. 78% (7 st) Yes.
2. 11% (1 st) Don't know.
3. 11% (1 st) No.
What is your level of education?
1. 56% (5 st) Master's student.
2. 22% (2 st) PhD student or PhD.
3. 22% (2 st) Other.
How did you hear about this course? (Please check all alternatives that apply.)
1. 22% (2 st) Word of mouth.
2. 0% (0 st) Lecturer on other course.
3. 11% (1 st) Message on mailing list.
4. 11% (1 st) Message sent out to students on my program.
5. 0% (0 st) Ad on bulletin board.
6. 33% (3 st) Ad on the web.
Which of those sources of information had a significant impact on your decision to attend this course? (Please check all alternatives that apply.)
1. 33% (3 st) Word of mouth.
2. 0% (0 st) Lecturer on other course.
3. 11% (1 st) Message on mailing list.
4. 11% (1 st) Message sent out to students on my program.
5. 11% (1 st) Ad on bulletin board.
6. 22% (2 st) Ad on the web.
Information and interaction
How often did you refer to the course webpages to find information about the course?
1. 22% (2 st) Several times per week.
2. 67% (6 st) Once or twice per week.
3. 11% (1 st) Just a few times per month.
4. 0% (0 st) Just a few times during the duration of the course.
5. 0% (0 st) Essentially did not refer to the webpages at all.
6. 0% (0 st) Not applicable.
What do you think about the quality of the course webpages? (Were they kept up to date? Did they contain the information you needed? Was the information easy to find?)
1. 22% (2 st) Very good.
2. 56% (5 st) Fairly good.
3. 22% (2 st) Acceptable.
4. 0% (0 st) Fairly bad.
5. 0% (0 st) Very bad.
6. 0% (0 st) Not applicable.
Any comments on the course webpages?
A mixed back, the use of Piazza was quite good, though somewhat annoying to need yet another site for courses. I would have preferred to not use the old CSC style website, and instead use KTH Social (which I anyway thought was a requirement for new course offerings). Since news was posted on Piazza, you still got email notifications.
---
The course literature section is not update, only Aurora-Barak is used.
It would benefit from mentioning that Arora-Barak can be accessed online through KTHB.
During the course we had a number of "opinion polls" with a few quick questions about the course that you were asked to answer. What do you think about these opinion polls?
1. 33% (3 st) They were really good and/or seemed very useful.
2. 33% (3 st) They were fairly good and/or seemed somewhat useful.
3. 11% (1 st) Neutral as to the value of these polls.
4. 11% (1 st) They were fairly bad and/or did not seem very useful.
5. 11% (1 st) They were really bad and/or seemed useless.
6. 0% (0 st) Not applicable.
Any comments on the opinion polls?
I enjoyed the fact that the lecturer/course responsible took incentive to modify the course using feedback while it was running, usually feedback is only given at the end for the next offering (like this).
---
There was some criticism in the optinion polls. Sometimes this didn't really affect the further course although the same critics appeared several times.
---
It would be preferable if they were online so one does not have to show up to the lecture in order to comment and influence the course content.
---
Other courses should learn from this one. This is very good.
We used Piazza for student-instructor interaction on the course. What did you think about Piazza (great, so-so, or bad)? Anything in particular that you liked or did not like? Any comparisons to KTH Social or other similar tools? What do you think in general about organizing student-instructor interaction in this way?
It didn't seem to add very much to the course.
---
I think Piazza worked quite well, though a bit spammy with the emails. Since there were points for answering problems, it would be dangerous to disable emails. I would have preferred to only use KTH Social, but I realise that its functionality may not be suited for those kinds of problems solving sessions. The Latex support was nice, though I think KTH Social also supports it to some extent.
---
Haven't used any similar tools previously. Did really like the openness and possibilities it created
---
I liked Piazza. Nice threaded organization of posts, and overview of old threads. I can't say I felt the non-standard LaTeX syntax or other things people were complaining about was a problem.
---
LaTeX on Piazza is great, Piazza does perfectly fit such courses.
---
I haven't used anything better and it works well with the peer-correction in the course.
---
Great, although I think you shouldn't need double dollar signs to invoke mathjax/latex…but that's the only downside I see…

I see no viable alternative to Piazza, so keep it if anything.

Lectures
How many of the lectures did you attend (23 all in all including guest lectures)?
1. 0% (0 st) Less than 20%.
2. 22% (2 st) 20-40%.
3. 0% (0 st) 40-60%.
4. 0% (0 st) 60-80%.
5. 78% (7 st) More than 80%.
What do you think about the number of lectures on the course?
1. 0% (0 st) Way too many.
2. 11% (1 st) A bit too many.
3. 56% (5 st) About right.
4. 22% (2 st) A bit too few.
5. 0% (0 st) Way too few.
6. 11% (1 st) Don't know — didn't attend lectures.
What do you think about the regular (non-guest) lectures by Jakob Nordström from a pedagogical point of view? (Was the material explained well? Did the lecturer speak and write clearly?)
1. 22% (2 st) Very good.
2. 44% (4 st) Good.
3. 22% (2 st) Acceptable.
4. 0% (0 st) Fairly bad.
5. 0% (0 st) Very bad.
6. 11% (1 st) Don't know — didn't attend lectures.
Any comments about the pedagogical contents of the regular lectures? (Bonus points for constructive criticism.)
I thought all lectures were clear and well prepared. The hand-outs of lecture notes were great as you didn't have to take notes as vigorously as you would otherwise need. Due to the complexity of the material (no pun intended), it was hard to follow along during lectures. The lecturer was engaged and tried to prod the students on whether they followed along or not, which was good. No immediate constructive criticism comes to mind, unfortunately.
---
No fault of the teacher but a room with more whiteboards wouldn't hurt for a few of the lectures. The lecture notes were really good and helped a lot.
---
The speed was always very tough and maybe too fast. Only when Jakob started to hand out the lecture notes in advance it was possible to follow. This should be done in the next course from the beginning.
---
Focus on too much technical detail instead of building understanding. Should probably have had a few more lectures to give you more time for this (it felt like you skipped saying things you wanted because of lack of time).
In particular, what do you think about the pace of the regular lectures?
1. 11% (1 st) Way too fast.
2. 33% (3 st) A bit too fast.
3. 33% (3 st) About right.
4. 22% (2 st) A bit too slow.
5. 0% (0 st) Way too slow.
6. 0% (0 st) Don't know — didn't attend lectures.
Any comments about the pace of the lectures?
Hard to say, the pace was quite high, but to cover all material it seemed to be needed. Also, I tend to have problems following along during lectures, I need to go through the material afterwards to understand it.
---
Some, more towards the end were quite fast.
---
I personnaly think that we crossed the line, where the course becomes too overloaded. On a longer perspective I would profit more from the course if we did less content, but this content in more detail.
What do you think about that the first three lectures recapitulated prerequisites from previous courses?
1. 33% (3 st) Recap should have been skipped completely.
2. 11% (1 st) Recap should have been faster.
3. 11% (1 st) Neutral/no opinion.
4. 22% (2 st) Recap was good.
5. 22% (2 st) Recap was absolutely essential.
6. 0% (0 st) Don't know — didn't attend these lectures.
Any comments or suggestions for improvements regarding the recap?
Personally I felt that it was good to recap older material, but I don't know if I would consider it a necessity.
---
As I lacked the pre knowledge from those courses the recap was essential for my understanding. Together with reading some on some other sources it helped me to be able to follow the course.
---
(suggesting to skip it entirely in the sense that I didn't personally need it; obviously this is not an actual suggestion since other people did)
---
Recap should be done in the same way.
How often did you read new material in the textbook or notes to prepare for an upcoming lecture?
1. 11% (1 st) Almost every lecture.
2. 33% (3 st) Roughly every second lecture.
3. 11% (1 st) Roughly every fourth lecture or so.
4. 11% (1 st) Maybe for a couple of lectures all in all.
5. 33% (3 st) Pretty much never.
6. 0% (0 st) Not applicable.
How often did you go over the textbook or notes to repeat the material covered in a lecture to refresh your memory before the next lecture?
1. 33% (3 st) Almost every lecture.
2. 22% (2 st) Roughly every second lecture.
3. 11% (1 st) Roughly every fourth lecture or so.
4. 11% (1 st) Maybe for a couple of lectures all in all.
5. 22% (2 st) Pretty much never.
6. 0% (0 st) Not applicable.
The course had a couple of guest lectures. In general, what do you think about having guest lectures on a course like this?
1. 44% (4 st) Very good — they should definitely be kept.
2. 44% (4 st) Good — it's reasonably valuable to have them.
3. 11% (1 st) Neutral — they didn't hurt but skipping them would probably be just as fine.
4. 0% (0 st) Fairly bad — we should have had less of them.
5. 0% (0 st) Very bad — guest lectures should have been skipped altogether.
6. 0% (0 st) Don't know — didn't attend the guest lectures.
Any comments or feedback about the guest lectures in general?
It was interesting to get some other people in on the subject, but I would not say it is as essential in a theoretical course. In a more practical course, it might be more important to get people from the industry in on the subject.
---
Great idea, should definitely be kept!!
---
Those were the best lectures in my opinion - they made a natural "pause" in the course, where one could focus more on "exploring" the subject (and surrounding such) of the course without focusing too much on the usual administrativia of homework, examination, deadlines and whatnot.
How did you like Johan Håstad's guest lectures?
1. 33% (3 st) Very good.
2. 22% (2 st) Good.
3. 22% (2 st) Acceptable.
4. 0% (0 st) Fairly bad.
5. 0% (0 st) Very bad.
6. 22% (2 st) Don't know — didn't attend the lectures.
Any comments or feedback to Johan?
Nothing comes to mind, good but hard to follow along without much time spent on preparation (personally).
---
Interesting material and nice pace!
---
The first lecture was great as it was. The second lecture was a far too abstract topic that did not suit the courses material. It is nice to know that such bounds exist, but there was neighter enough time to explain the methods used in proofs nor to properly understand the statements. It would be great to do a lecture during the lectures on the PCP-theorem about the approximability of 3SAT.
How did you like Danupon Nanongkai's guest lectures?
1. 56% (5 st) Very good.
2. 11% (1 st) Good.
3. 22% (2 st) Acceptable.
4. 0% (0 st) Fairly bad.
5. 0% (0 st) Very bad.
6. 11% (1 st) Don't know — didn't attend the lectures.
Any comments or feedback to Danupon?
Quite good use of power points for illustrations, though it would have been nice if the slides contained more information on them for later reference.
---
Interesting subject and results, but I felt the pace was a bit too slow, and I would have preferred blackboard lectures rather than slides. (Thank you for guest lecturing!)
---
Great lecturer! Keep going like this. The transmission of intuition is great and the pictures help a lot to understand the content.
---
Those lectures inspired me to set out to learn about distributed computing (which I believe is the best possible outcome one could possibly have from a lecture)
What do you think about the choice of topics on the course?
1. 22% (2 st) Very good.
2. 56% (5 st) Good.
3. 22% (2 st) Acceptable.
4. 0% (0 st) Fairly bad.
5. 0% (0 st) Very bad.
6. 0% (0 st) No opinion.
Any particular comments on the choice of topics? Should we have had more of something? Less of something? Something that should have been included that we totally missed?
I am not informed enough in the field to have preferences outside what the course covered, but I found the topics to be interesting.
---
The overview of advanced topics was source of good inspiration for new topics to read about.
---
All topics were interesting, but there were too many topics.
---
Felt like a fairly standard course in complexity theory.
---
Even a short overview of quantum computing would have been fun. Other than that, perhaps the complexity of counting (e.g. the class #P and so on) could have been fun to learn about. I think the recap wasted a little time that could have been used to do a little more new stuff (it was still good to have the recap though). But then again, I can read up on other topics on my own…I don't need a formal course to learn about stuff
What do you think about that we have (very intentionally) tended to run over 5 minutes on the first half of the lectures and only get a 10-minute break?
1. 22% (2 st) Very good; makes excellent sense to use the time this way.
2. 33% (3 st) Fairly good; seems to make sense to use the time this way.
3. 33% (3 st) OK/neutral as to whether this is good or bad.
4. 0% (0 st) Fairly bad; probably would have been better to have a 15-minute break.
5. 0% (0 st) Very bad; definitely would have been better to have a 15-minute break.
6. 11% (1 st) No opinion.
What do you think about that we sometimes (less intentionally) ran over on the second half of the lectures and finished a bit late?
1. 11% (1 st) Very good; makes excellent sense to use the time this way.
2. 11% (1 st) Fairly good; seems to make sense to use the time this way.
3. 33% (3 st) OK/neutral as to whether this is good or bad.
4. 33% (3 st) Fairly bad; lectures should preferably end on time.
5. 0% (0 st) Very bad; lectures should definitely end on time.
6. 11% (1 st) No opinion.
Problem sets
Did you personally find the problem sets easy or hard?
1. 0% (0 st) Very easy.
2. 0% (0 st) Fairly easy.
3. 22% (2 st) Medium.
4. 33% (3 st) Fairly hard.
5. 44% (4 st) Very hard.
6. 0% (0 st) Don't know — didn't do problem sets.
In general, how would you assess the appropriateness of the level of difficulty of the problem sets?
1. 0% (0 st) Far too easy.
2. 11% (1 st) A bit too easy.
3. 33% (3 st) About right.
4. 56% (5 st) A bit too hard.
5. 0% (0 st) Far too hard.
6. 0% (0 st) Don't know — didn't do problem sets.
What do you think about the number of problem sets and the number of problems per set?
1. 0% (0 st) Far too few problems all in all.
2. 0% (0 st) A bit too few problems.
3. 56% (5 st) About right.
4. 33% (3 st) A bit too many.
5. 11% (1 st) Far too many.
6. 0% (0 st) Don't know — didn't do problem sets.
What do you think about the deadlines for the problem sets and how much time was given for them?
1. 0% (0 st) Far too much time.
2. 11% (1 st) A bit too much time.
3. 78% (7 st) About right.
4. 11% (1 st) A bit too little time.
5. 0% (0 st) Far too little time.
6. 0% (0 st) Don't know — didn't do problem sets.
What do you think about the grading of the problem sets?
1. 11% (1 st) Far too harsh.
2. 44% (4 st) A bit too harsh.
3. 33% (3 st) About right.
4. 0% (0 st) A bit too lenient.
5. 0% (0 st) Far too lenient.
6. 0% (0 st) Not applicable.
How much time on average would you say that you spent per problem set?
1. 0% (0 st) Less than 5 hours.
2. 0% (0 st) 5-10 hours.
3. 33% (3 st) 10-20 hours (up to half a week of working time).
4. 44% (4 st) 20-30 hours.
5. 11% (1 st) 30-40 hours (up to a full week of working time).
6. 11% (1 st) More than 40 hours.
7. 0% (0 st) Not applicable.
How well do you think the problem sets corresponded to the material presented during the lectures?
1. 44% (4 st) Very well.
2. 56% (5 st) Fairly well.
3. 0% (0 st) Somewhat, but not too well.
4. 0% (0 st) Fairly badly.
5. 0% (0 st) Very badly.
6. 0% (0 st) Don't know — didn't do problem sets.
Do you have any comments or feedback on the problem sets?
They were mostly very interesting and fun.
---
The last problem sets felt a bit too hard, and I think it was crucial that collaboration was allowed. I did not keep track of time spent, so I cannot really tell how many hours on average I spent per problem set (a significant amount). They grading was quite harsh, but I understand the need for mathematical precision in presenting solutions.
---
About the problem sets:
I liked the problems, on the whole. It was a nice mix of understanding definitions, working with them a bit more in depth, and creative problem solving.

About the grading:
It was harsh, and of varying harshness between problems in an unpredictable manner. Sometimes I got 0 points on problems I still feel are rigorous and correct, and full points on other problems where I was quite sloppy or not comprehensive. I can accept some points taken away by bad explanations, but what "easy to understand" means varies from person to person and is hard to optimize for on your own. Hence I feel such things should have been given less weight.

I also feel that given the large amount of problems, explicitly allowing terser solutions would be nice.

The time consumed by the homework (in large part due to the harsh grading) is the reason I can't blindly recommend other people to take the course in the future; instead I have to add a waiver that apart from just being a generally hard course it will also consume all their time and they probably shouldn't take it at the same time as other heavy courses. Of course this is counter-balanced by people learning things more deeply, but it's worth paying serious attention to.
---
I think, that there should be more problem sets, e.g. 8 problem sets with half the number of problems.
---
The grading was horribly harsh. There was an absurd focus on technical detail. I don't mind being very picky about these details in general, but the point deductions seemed very disproportionate.

In general, I really enjoyed the problems though. They are probably my second favorite at KTH, after the crypto course.
---
It might be an idea to be able get full-score without finishing all exercises. Some peer-corrections took an indecent amount of time trying to figure out what the author meant due to way too hastily written solutions. It seems the course grading encourages attempting to solve as much as possible instead of using new vocabulary and learning how to write correct and understandable proofs.
---
The only real thing I have against it is that it was very hard to predict how harsh the actual grading would be, and what the grader would look for, after handing in the psets. Things sort of degenerated into a "it doesn't really matter how well I think I have done on this pset, it'll still be heavily deducted due to some arbitrary reason the grader will say I've clearly missed"-kind of situation. It was disheartening to see what seemed like a solid solution get massacred for seemingly arbitrary reasons. It is of course hard to better clarify what the grader looks for in a solution without sacrificing the difficulty of the problems, so I really don't have too much of a suggestion for improvement on this.

Perhaps the grader(s) could help (more) in pointing out what characterizes a good solution during the pset discussions by, during the phase where everyone gets their bonus points, point out extra good / bad parts of the posted solutions, and hopefully also why those parts are good?

Then again, that might be a bit much to ask, especially on an advanced / phd level course, and considering how busy the grader(s) are in general, and since every submitted solution has its own good / bad sides

Peer evaluation
Did you personally find it a useful exercise to peer evaluate the problem set solutions of a fellow student? Did you learn anything from doing the peer evaluation?
1. 11% (1 st) Very useful.
2. 22% (2 st) Fairly useful.
3. 33% (3 st) Neutral.
4. 33% (3 st) Not very useful.
5. 0% (0 st) Not at all useful.
6. 0% (0 st) Not applicable; didn't participate in peer evaluations.
Do you think that having peer evaluation as one of the requirements is a relevant form of examination?
1. 33% (3 st) Very relevant.
2. 22% (2 st) Fairly relevant.
3. 0% (0 st) Neutral.
4. 33% (3 st) Not very relevant.
5. 11% (1 st) Not at all relevant.
6. 0% (0 st) Not applicable; didn't participate in peer evaluations.
Did you personally receive useful feedback on your problem set solutions from your peer evaluator? (Note that this is asking about the peer evaluation comments, not about the instructor's grading.)
1. 11% (1 st) Very useful.
2. 22% (2 st) Fairly useful.
3. 22% (2 st) Neutral.
4. 33% (3 st) Not very useful.
5. 11% (1 st) Not at all useful.
6. 0% (0 st) Not applicable; didn't participate in peer evaluations.
Did you personally find the posting and discussion of solutions on Piazza useful? Did you learn anything from it?
1. 33% (3 st) Very useful.
2. 11% (1 st) Fairly useful.
3. 22% (2 st) Neutral.
4. 22% (2 st) Not very useful.
5. 11% (1 st) Not at all useful.
6. 0% (0 st) Not applicable; didn't participate in peer evaluations.
Do you have any comments or feedback regarding the questions above or on any other aspect of the peer evaluation process?
I generally dislike peer evaluations, especially if they are "graded", as I think that should be done by the teachers for a fair examination. Though I found it quite useful to read discussions on Piazza afterwards. I guess it is technically useful to see other student's solutions.
---
I like the bonus point idea of piazza but the "rush to be first" created some problems. Instead maybe leave it open and hand out points between all who contributed with something on that problem and not give points for the posting of a correct solution.
---
Re Piazza: I didn't much like that the system gave lots of bonus points to the first person to copy-paste their solution to Piazza. I'm not sure how to organize it better in practice, but perhaps it'd be nicer to have something that rewards people who didn't solve, but instead read, understood and explained someone else's solution. On the up-side, there quickly appeared solutions to everything, and reading through people's solutions and discussing things was fun and useful. Maybe decrease the amount of bonus points given for posting solutions and/or make them less proportional to the points given on the homeworks?
---
The system doesn't really lead to what it should. It is intended that students rethink the exercises. However, usually there isn't much participation on the discussions and there is only a run on posting solutions as fast as possible and stopping to comment the same day.
---
I liked the concept of posting and discussing solutions on Piazza, but I personally think the bonus point concept needs reworking. It would be nice to have a fixed time for when the solution posts were added.

I wouldn't have needed bonus points as an incentive for posting my solutions and discussing them, so for me they made no sense -- I'd rather have had a bit lower grading limits. But maybe other people were motivated by them.
---
With the "storm of pset3" in mind, perhaps one could have better structure on the bonus point rules than "first come, first served". However, I have no suggestions to add here.

Paper presentation
Do you think that having an oral presentation of a paper as one of the requirements for the highest grade on the MSc course and for a pass on the PhD course is a relevant form of examination?
1. 11% (1 st) Very relevant.
2. 44% (4 st) Fairly relevant.
3. 22% (2 st) Neutral.
4. 11% (1 st) Not very relevant.
5. 0% (0 st) Not at all relevant.
6. 11% (1 st) Not applicable; will not present a paper.
Do you have any comments or feedback regarding the paper presentation requirement?
I did not make a presentation, but it can be healthy to be forced to delve into a proper paper once in a while.
---
Still haven't done mine. Preparing it is certainly more time-consuming than I expected. Scheduling a fixed date with clear expectations about what should be done by when, and giving a link to some example presentation, would be good I think.
---
I would rather it remains (from a MSc point of view)
---
It could have been fun if you could do a paper presentation (almost) regardless of pset-grades, and maybe get bonus points instead, to hopefully raise the grade that way upon a well done presentation. Then again, papers are tough to read unless you know the course material well

Workload and course credits
How much of your total study/working time did you spend on this course (on average per week, say)?
1. 11% (1 st) Less than 15%.
2. 11% (1 st) 15-30%.
3. 44% (4 st) 30-50%.
4. 22% (2 st) 50-70%.
5. 11% (1 st) More than 70%.
This course gives 7.5 ECTS credits. What you you think about this compared to the number of credits given for other courses?
1. 56% (5 st) Should have given more than 7.5 credits.
2. 33% (3 st) 7.5 credits was about right.
3. 11% (1 st) Should have given less credits.
Concluding questions
Did you experience that you were discriminated against on this course due to gender, sexuality, ethnicity, or disability?
1. 0% (0 st) Yes.
2. 0% (0 st) Don't know.
3. 100% (9 st) No.
If yes, in which way?
How do you assess the course from a gender perspective (e.g., with respect to course material, course contents, lecturer et cetera)?
I don't think there was anything wrong with the course material/content, which from what I remember seemed to be pretty gender neutral. Of course all (even if there were only three) lecturers and all except one student (I think) were men, which is disappointing.

I don't think the course itself can be faulted, but it's obvious KTH and CSC in general have a large problem with gender, just by looking at the attending students in any class.
---
Nothing problematic comes to mind, I can't recall if the lecturer and course book used gender neutral pronouns or not when referring to hypothetical persons.
---
Good / no discrimination.
---
N/A
---
Not applicable
---
Neutral
Although we did not do so this time, the course is meant to be given with ca 16-20 45-minute lectures (or possibly ca 8-10 90-minute lectures), not ca 20 90-minute lectures. It should still cover essentially the same amount of material, but the students should do a larger portion of self-study. How do you think such a set-up would have affected this course? What could/would have been positive and/or negative aspects?
A lot of self-study was still needed during this course, more than in pretty much any other course I have taken. Having even less lectures would, for me personally, be mostly negative, as I feel I learn better by combining going to lectures and reading the material, at least in such a hard course. Even though I didn't really prepare for most lectures, I feel they were very valuable since I gained a lot of intuition from the instructors explanations, and reading the material afterwards became easier. If others feel differently there is always the option of skipping lectures.
---
I don't think it would have helped to have less lecture time, the material was heavy and any help given was appreciated.
---
That would increase the difficulty by a lot. For me that would make the course load too much and I would probably not have completed the course.
---
This would make the course a more classic one where the lectures build upon each other. Now there were several independent topics.

I think such course would be better for master students that do not do their thesis in TCS.
---
Really hard for me to say; I don't really have experience with courses requiring self-study.
---
it should be exactly contrary, more lectures are needed if that much material is inteded to be covered.
---
That seems way worse than what we had now. On the contrary, I think there should have been more lecture time than now.
---
No comment.
Do you have any comments on (general) aspects of the course that should stay the same in the next offering of the course?
The lectures were mostly really good, and the problem sets were all interesting and engaging. They should stay roughly the same.
---
The examination with problem sets was good. Hand-outs during lectures was also good. I liked the structure of lectures and structure of the course in general. Good communication trough Piazza.
---
I think the course was very good and really enjoyed it. I recommend it to a lot of people so I don't think there is any reason to change the format. The amount of lectures and problem sets were very good
---
Most of it. I liked the general setup and the course material.
---
Good level and selection of problems on problem sets.
Do you have any suggestions for how the course could be improved?
The first-come-first-serve basis for distributing points for submitting solutions was really bad, and should be changed in some way. I was only able to take part in the process for the first problem set, which of course is the only one you don't really need any extra points for. For all the others I had other lectures or labs to attend to.
---
I would have preferred to not have peer review as part of the examination. I think the last problem sets could be made a little bit less hard. It would be nice with Latex:ed lecture notes for readability.
---
Maybe we should change the venue to Djuronaset, say.
---
Basically the things noted previously: slightly lighter workload for the homework would I think be good, and less unpredictable grading.
---
More problem sheets with each less problems.
---
Bonus point system; # of lectures (more); less harsh grading
Any other final comments on the course?
It was one of the interesting courses I've taken at KTH.
---
I found the course to be highly interesting. This is probably one of the toughest courses I have taken at KTH due to the level of the material and how rigorous the grading was, but I learned lots.
---
Great course, thank you! It's one of the few times I've really felt that I've learnt things from a course.
---
Although I did never spend that much time on a course and it was overloaded, I can say, that it was absolutely worth taking the course. It will also affect my further studies in the direction of complexity theory.
---
Really nice course. One of the most interesting I've studied at KTH.

Course Evaluation for DD2445 Complexity Theory, Autumn 2015

General questions about the course

Information and interaction

Lectures

Problem sets

Peer evaluation

Paper presentation

Workload and course credits

Concluding questions