Visa resultat

Stochastic optimization algorithms, FFR105

Teaching

1. How do you rate the lectures?

Not good at all»		0		0%
Not so good»		0		0%
Neither good nor bad»		1		5%
Good»		10		55%
Very good»		7		38%

Genomsnitt: 4.33

- Sometimes all that was said in the lectures already stood in the book, word for word, so it wasn"t really interesting at that time. Try involving more fun examples. » (Good)
- Lectures were just done in a bit fast manner.» (Good)
- the lectures are verly clere for the most paret. but some of the proves can be dificelt to folow.» (Good)
- Mattias was always very well organized» (Good)
- Alltid intressant och strukturerat, men ibland för stressigt.» (Very good)
- Perhaps not the most glamorous lectures, but very clear and well structured, like the rest of the course.» (Very good)
- Mattias is very pedagogical and explains everything at a good speed. The fact that what he says on the lectures almost is exactly what’,s written in the book makes it very easy to catch up a missed lecture. If one is to say something that needs to be improved it is that he gets very annoyed if someone is late. That has sometimes made students skip the first part of the lecture since they know they will be late. I don"t think neither Mattias nor the students want that to be the case.» (Very good)

2. To what extent have the lectures been of help for your learning?

Not at all»		0		0%
Some extent»		3		16%
Large extent»		13		72%
Very large extent»		2		11%

Genomsnitt: 2.94

- since the lectures are about whats in the book oc» (Large extent)
- For me lectures are a way to make me study. They have been successful in that way.» (Large extent)

3. Do you think that the lectures complemented the course literature well?

No, not at all»		0		0%
Not so much»		3		17%
Yes»		11		64%
Yes, very much»		3		17%

Genomsnitt: 3

- other waý round, i"d say» (?)
- Seems like the book and the lectures are more or less the same thing.» (Not so much)
- the lectures was pretty much the book side by side. I would appreciate if the lecture was more deep than the book. » (Not so much)
- Ibland lite för likt boken. Kanske hitta på andra exempel till lektionerna?» (Yes)
- Since they basically say the same thing, it makes studying easier, but at the same time you only get the one perspective.» (Yes)
- However, it would have been nice if the lectures provided extra material that was not available in the book» (Yes)
- see above» (Yes, very much)
- The content of the lectures is (almost) the same as in the book, which allows the students to freely choose the way in which they like best to study.» (Yes, very much)

4. How do you rate the exercise classes?

Not good at all»		1		5%
Not so good»		2		11%
Neither good nor bad»		7		38%
Good»		7		38%
Very good»		1		5%

Genomsnitt: 3.27

- Should be mandatory exercises (problem sets) for submission so that we could practically» (Not so good)
- I didn"t have time visit them. (I"m usually a bit to busy for my own good)» (Neither good nor bad)
- Actually managed to miss all of them, so I can"t tell» (Neither good nor bad)
- I appreciate david"s help. The classes were more helpful when he skipped over details such as algebra.» (Neither good nor bad)
- ok, not great but ok. » (Good)
- introduce live programming» (Good)
- They were OK, they filled their purpose well. I understand that it"s a bit hard to have exercise classes when there"s a very limited number of exercises which can be done analytically.» (Good)

5. To what extent have the exercise classes been of help for your learning?

Small extent»		4		22%
Some extent»		9		50%
Large extent»		4		22%
Very large extent»		1		5%

Genomsnitt: 2.11

- Same reason as above» (Small extent)
- Good preparation for the problem-solving parts of the exam.» (Some extent)
- Don"t know. As I said I didn"t visit them.» (Some extent)
- they went throu the problems ok, but overe all you didn"t get very mutsh out of them» (Some extent)
- It was helpful to work through problems, to see all the steps» (Large extent)

6. To what extent have the home problems been of help for your learning?

Small extent»		0		0%
Some extent»		0		0%
Large extent»		5		27%
Very large extent»		13		72%

Genomsnitt: 3.72

- in learning ACO, EA and PSO!» (Very large extent)
- very good problems all of them! i like that you receive point instead of only pass/retry. you put more effort in finishing with good result. » (Very large extent)
- I am a firm believer in "learning by doing".» (Very large extent)
- The home problems have been very interesting, which is the best way to encourage learning. They have also made one go through the important parts of the course. Very good!» (Very large extent)
- very useful to fuly anderstand the methodes» (Very large extent)
- Completing the homework really took a lot of hard work, but now I feel like I can apply the methods to other problems as well» (Very large extent)

7. What is your impression of the course literature?

18 svarande

Very poor»		0		0%
Poor»		0		0%
Average»		2		11%
Good»		6		33%
Very good»		10		55%

Genomsnitt: 4.44

- Confusing at times, some misprints which is understandable as the book is brand new. Parts I found particularly confusing are: Algorithm 4.1 (e.g. does not say that p must not be calculated for nodes in the tabu list) 5.4.1 (does not say that positions must NOT be truncated, I would rather say it does say that they should!) Also, 3.2.2.2 is very interconnected with B.2, and one must flip back and forth a lot to understand everything.In my opinion, a text should be possible to understand without reading any appendix in parallel, why not put everything related under the same section instead? Although it has nothing to do with the quality of the book itself, and surely not the fault of Mattias or anyone else related to the course, I must nevertheless point out that I found the book VERY expensive, at I think, 620 SEK! Surely the most expensive course book in cost/page i have ever had, which is too bad.» (Average)
- Vissa bevis skulle behöva fler förklarande steg.» (Good)
- good, clerly market out all importent algorithums and equashens» (Good)
- helps you understand the algorithms, sometimes it"s hard to read because of the small spacings in between equations and so forth» (Very good)
- very good book.» (Very good)
- Very good as course literature. A bit brief if you were to read it outside the context of the course, but that"s not the purpose either.» (Very good)
- The best course book I have had in all my years at Chalmers! The only course book I have read that can rival is a book by Björn Lantz used in his course at the school of business, economics and law at Gothenburg University (Handels). What those books have in common and what I think is very important is that they both are rather short and concise. Both books also have good examples that clarify what"s written in the text as well as clearly distinguished algorithms. The only negative aspect is the prize, but it would be worth the double, easily!» (Very good)
- I wish all my course literature was like this. The text is very efficient and says exactly just what is needed to be said, not to much not to little. They lay out is very good and makes it easy to read and find what one need.It was worth the money, even though it seemed a bit expensive at first.» (Very good)
- There are a few areas that could use clarification, but overall, very good» (Very good)

8. How well did the course administration, web page, communication etc. work?

Very badly»		0		0%
Rather badly»		0		0%
Rather well»		5		27%
Very well»		13		72%

Genomsnitt: 3.72

- prefer handing in stuff through the fire system» (Rather well)
- lectures notes should be upload in advance.» (Rather well)
- The course page could if one should be picky be improved quite a bit, but all the important information is there.» (Rather well)
- The homepage is not pretty but user friendly and anything one needed to know one could easily find. Could have been good to tell which aids one was allowed to use at the exam on the actual homepage. » (Very well)

Your own efforts

9. How many hours per week did you spend on this course?

17 svarande

At most 15 hours»		3		17%
About 20 hours»		7		41%
About 25 hours»		2		11%
At least 30 hours»		5		29%

Genomsnitt: 2.52

- this course is nothing without practice and for doing some practice problems set or problem exercises should be mandatory.» (At most 15 hours)
- It varies quite a bit depending on the hand ins, but I don"t think I needed to put down to much effort since it was so interesting and all information therefore went straight in.» (At most 15 hours)
- mostly worked to the end of the coures for sheet 2 otherwis ferly even» (About 20 hours)
- The closer to the days of hand in the more work was done.» (About 20 hours)
- The programming took a lot of time for me» (At least 30 hours)

10. How large part of the teaching offered (lectures and exercise classes) did you attend?

18 svarande

Less than 20%»		0		0%
20-40%»		1		5%
41-60%»		2		11%
61-80%»		1		5%
More than 80%»		14		77%

Genomsnitt: 4.55

- I attended the lectures that I could, but as I previously wrote I may be a bit to busy for my own good.» (20-40%)

Study climate

11. How were the opportunities for asking questions and getting help during the course?

Very poor»		0		0%
Rather poor»		1		5%
Rather good»		0		0%
Very good»		10		55%
I did not seek help»		7		38%

Genomsnitt: 4.27

- No exercise/laboration classes where we could sit and work in a classroom with access to a supervisor. see below.» (Rather poor)
- Mattias was very helpful when I approached him » (Very good)
- I sadly enough have had little training in asking questions and it"s and it"s hard to break old habits » (I did not seek help)

12. How was the course workload?

Too low»		1		5%
Low»		0		0%
Adequate»		7		38%
High»		9		50%
Too high»		1		5%

Genomsnitt: 3.5

- work load is depend upon projects and assignments not just on final exam.» (Too low)
- It was quite good actually. You learnt much without needing to kill yourself on the way.» (Adequate)
- But together with the Neural Network class which demanded a lot of time and energy it kind of got to much. But that"s NN problem. » (Adequate)
- considering the work load in the other course was very high, this was also a bit high. If you take this course with an easier course then the work load would be adequate. » (High)
- Utmanande men överkomligt.» (High)
- ok but problematic wene bothe SOA and AAN heve big handines at the same time» (High)
- More "frustrating" than "high" perhaps... See below.» (High)
- I feel the expectations were reasonable» (High)

Course goals, level of difficulty, exam

13. How understandable are the course goals?

18 svarande

I have not read the goals»		3		16%
The goals are difficult to understand»		0		0%
The goals give some guidance, but could be clearer»		2		11%
The goals clearly describe what I am supposed to learn»		13		72%

Genomsnitt: 3.38

- but the goals are understood during what is focused in the lectures and in the book» (I have not read the goals)
- I generally think that course goals aren"t any help during the course. They are however very helpful when deciding which courses to take. I don"t have any real opinion in the formulations of the goals in this course.» (The goals clearly describe what I am supposed to learn)
- Mattias is very careful about specifying expectations» (The goals clearly describe what I am supposed to learn)

14. Is the level of difficulty of the course reasonable, considering your background and the number of credits?

No, the level is too low»		0		0%
Yes, the level is reasonable»		17		94%
No, the level is too high»		1		5%

Genomsnitt: 2.05

- however, some additional resources for programming in MatLab would be great!» (Yes, the level is reasonable)
- in starting lectures basic concepts should be explain with practical exercises so that all course could be easily manage.» (No, the level is too high)

15. Did the examination as a whole assess whether you have reached the goals?

16 svarande

No, not at all»		0		0%
Yes, to some extent»		4		25%
Yes, definitely»		12		75%

Genomsnitt: 2.75 (bidrar till totalt genomsnitt/jämförelseindex)

- Har inte läst målen...» (Yes, to some extent)
- Doubtful if it assessed whether to decide which model to use for a given problem. Also, but less importantly, it did not test the knowledge of the biological processes that inspired the algorithms.» (Yes, to some extent)
- yes I think so» (Yes, definitely)
- The home problems test if you can implement the algorithms and the exam tests if you have understood the algorithms without reading from the book. What one can say about the exam is that many of the "proofs" aren"t very proofish and therefore hard to remember. I would personally think that some of the "proofs" only should be needed to know about so you could concentrate more on them with theoretical value for the course.» (Yes, definitely)
Genomsnitt totalt för detta stycke: 2.75

Summarizing questions

16. What is your general impression of the course?

Very poor»		0		0%
Poor»		1		5%
Average»		0		0%
Good»		7		38%
Very Good»		10		55%

Genomsnitt: 4.44

- See below.» (Poor)
- good clere and useful» (Good)
- one of the best so far (my fourth year)» (Very Good)
- En av de roligaste, intressantaste och utmanande kurser jag läst.» (Very Good)
- One of the best courses this far on Chalmers!» (Very Good)
- The things we learned were very interesting, and the applications we learned about seem very exciting (robotics, volvo, etc)» (Very Good)

17. Is there anything you think should be changed until next year?

18. Which areas do you think need more emphasis (or less emphasis) in the course?

- thinking, solve logical problems. if some one can make a logic to solve a problem set then coding or writing pseudo code is nothing.»
- More emphasis on the ACO, initiating pheromones and perhaps more discussions on comparing all the methods»
- Stochastic optimization needs more emphasis.»
- The balance was good. I also appreciated the biological backgrounds.»
- Some proofs maybe. The other things are needed.»
- PSO hade very litel prakticle and also ACO. may be ot mutsh enfeses on EA. LGP could have needed just a litel more»
- Classical optimization needs less attention, as I suppose many already are familiar with it.»
- Less on artificial neural networks and perhaps a little less on classical optimization, and more on the core of the course.»
- I think it was good as it was. Especially like the lectures which handled applications. It"s always good to be able to connect with reality.»
- More emphasis: programming, applications Less emphasis: classical optimization, proofs»

19. Additional comments