Coursera/Stanford online algorithms I course – a retrospective • Hacker News Board

Re: Coursera/Stanford online algorithms I course – a retrospective

#31

Post by DennisP » Wed, May 09, 2012, 12:11 AM UTC

otoburb wrote: post #29

gms wrote: post #23

Earlier quoted context omitted.

a million upvotes this class is killing me

The beauty of online courses is that people will hopefully have the opportunity to cheaply repeat lectures ad-nauseum until they finally feel they really KNOW their stuff. This is contrasted to the brutal method of being forced to learn things within a short period of time, often forgetting what you learned or just picking up banal mechanics. The focus shifts back to thoroughly learning course materials. Online educa…

That's exactly my plan for PGM. If I'd had nothing else keeping me busy maybe I would've kept up, but as soon as I fell behind it was clear I was in trouble.

Coursera's talked about maybe taking videos down for a while between courses, I hope they don't. I'm going to save them just in case, but it'd be nice to have the inline quizzes.

Re: Coursera/Stanford online algorithms I course – a retrospective

#32

Post by icegreentea » Wed, May 09, 2012, 1:18 AM UTC

singular wrote: post #28

Perhaps I've been out of formal education for too long (esp. on the mathematical side of things), but I did find the material quite challenging, more so the problem sets than the programming challenges. Part of that was emotional, I didn't do a computer science degree and thus find this kind of stuff almost... intimidating. It's the sort of stuff that gets asked at 'serious' interviews and thus I've had a sense of it…

I'm gonna back you up here, though from a slightly different perspective. People like you are exactly why this course seems easy for some other people. As what is essentially an introductory course for this subject, it should not be aiming to really challenge the top quartile of the class (or whatever). It should aim to provide the best possible foundation for the next course for as many people as reasonable.

The fact that the course has managed to reach out to all sorts of people (including someone like you) is something that should be celebrated. That it was designed so that someone who isn't 'good at math' (whatever that means) struggled a bit, but still got through it all is amazing. That's something that really makes a great introductory course.

After all, it's stupid easy to make a course hard, or to assign shit loads of work. The real trick is making it just hard enough.

Re: Coursera/Stanford online algorithms I course – a retrospective

#33

Post by warmfuzzykitten » Wed, May 09, 2012, 1:40 AM UTC

Homunculiheaded wrote: post #21

It's funny how often people complain "too easy" but I've never heard someone complain "didn't learn enough", in regards to the coursera courses. I think it's an indicator of how deeply broken our education system is that when assessing the value of a class people worry about how smart it made them feel (this is after all what it really means to pass a 'difficult' class), rather than how much they learned and could do…

Online course reviews written by current university students should be taken with a grain of salt. "Easy" has a special meaning to bright university students; it means "too many people would pass."

I once had a professor introduce me to a colleague as "He was the A." He seemed to think the remark praised him as much as me. Only in primary schools do you hear a few voices saying "90% of the class should be performing at A level".

I note that this reviewer didn't get full marks (perfect score) in the class. It's interesting to contrast the review, also on HN, from one who did. http://henrikwarne.com/2012/05/08/coursera-algorithms-course...

I read another review by a current Stanford student who complained that the pre-Coursera Machine Learning class was too easy. I wonder how he did? A close friend of mine took that course and earned a perfect score. He didn't complain it was too easy; he went back to his company and started a group to apply ML to their business. Easy/hard is beside the point. The point is: students learn.

Re: Coursera/Stanford online algorithms I course – a retrospective

#34

Post by wglb » Wed, May 09, 2012, 1:41 AM UTC

singular wrote: post #28

Perhaps I've been out of formal education for too long (esp. on the mathematical side of things), but I did find the material quite challenging, more so the problem sets than the programming challenges. Part of that was emotional, I didn't do a computer science degree and thus find this kind of stuff almost... intimidating. It's the sort of stuff that gets asked at 'serious' interviews and thus I've had a sense of it…

I didn't read it that way at all. Instead of seeing oh it was all too easy i heard I wish it had been harder; and I didn't see any competitive thread in it at all.

Re: Coursera/Stanford online algorithms I course – a retrospective

#35

Post by sesqu » Wed, May 09, 2012, 4:45 AM UTC

aplusbi wrote: post #26

lawnchair_larry wrote: post #11

Earlier quoted context omitted.

You shouldn't be grading then. They are not taking a class in Paying Attention, they are taking a class in algorithms. If they know the material, that is what matters. I also cannot stand this practice.

And people who know the material should be able to answer those questions without any difficulty. What's the problem?

I'm assuming the problem is questions like this:

  Recall the Master Method and its three parameters a,b,d. Which of the following is the best interpretation of b^d, in the context of divide-and-conquer algorithms?

Alternatively, the problem might be questions that ask for the state of an algorithm at some point, when the algorithm can be correctly implemented in different ways and one is expected to recall/go back and check the exact implementation used.

Re: Coursera/Stanford online algorithms I course – a retrospective

#36

Post by singular » Wed, May 09, 2012, 9:24 AM UTC

icegreentea wrote: post #32

singular wrote: post #28

Perhaps I've been out of formal education for too long (esp. on the mathematical side of things), but I did find the material quite challenging, more so the problem sets than the programming challenges. Part of that was emotional, I didn't do a computer science degree and thus find this kind of stuff almost... intimidating. It's the sort of stuff that gets asked at 'serious' interviews and thus I've had a sense of it…

I'm gonna back you up here, though from a slightly different perspective. People like you are exactly why this course seems easy for some other people. As what is essentially an introductory course for this subject, it should not be aiming to really challenge the top quartile of the class (or whatever). It should aim to provide the best possible foundation for the next course for as many people as reasonable. The fac…

Not to make this too much about ego :-) but I am actually quite good at maths, did v. well at school, did engineering degree, etc. so I'm not sure I'm a great example, rather I am out of practice and not as comfortable with this stuff now as I once was :-)

Re: Coursera/Stanford online algorithms I course – a retrospective

#37

Post by aplusbi » Wed, May 09, 2012, 2:39 PM UTC

sesqu wrote: post #35

aplusbi wrote: post #26

Earlier quoted context omitted.

And people who know the material should be able to answer those questions without any difficulty. What's the problem?

I'm assuming the problem is questions like this: Recall the Master Method and its three parameters a,b,d. Which of the following is the best interpretation of b^d, in the context of divide-and-conquer algorithms? Alternatively, the problem might be questions that ask for the state of an algorithm at some point, when the algorithm can be correctly implemented in different ways and one is expected to recall/go back and…

Those kind of questions should be taken out back and put out of their misery, but I don't think there's anything to suggest that adestefan was asking those kinds of questions.

Re: Coursera/Stanford online algorithms I course – a retrospective

#38

Post by gms » Wed, May 09, 2012, 9:12 PM UTC

otoburb wrote: post #29

gms wrote: post #23

Earlier quoted context omitted.

a million upvotes this class is killing me

The beauty of online courses is that people will hopefully have the opportunity to cheaply repeat lectures ad-nauseum until they finally feel they really KNOW their stuff. This is contrasted to the brutal method of being forced to learn things within a short period of time, often forgetting what you learned or just picking up banal mechanics. The focus shifts back to thoroughly learning course materials. Online educa…

The phenomenon you describe is still ever-present in the online format if you stick to their deadlines. If you ignore them and the associated grade penalties, then ok I agree.

Re: Coursera/Stanford online algorithms I course – a retrospective

#39

Post by gfodor » Thu, May 10, 2012, 6:48 AM UTC

DennisP wrote: post #15

If you're looking for a hard class, Probabilistic Graphical Models might suit better. The staff says that other than a small portion of the homeworks which they couldn't auto-grade, it's the same material as the real course at Stanford, where it's one of the hardest classes in the graduate CS department.

Yeah, after taking ML class I was worried the online stuff was going to be heavily watered down. PGM class has reassured me that even after college I can still feel that "oh shit" moment opening up a problem set and realizing I am in way over my head. Still have my head above water though!