Here you go: http://www.ai-class.com/
They're still working out kinks in the site, and no homework is up yet, but the initial set of lectures (really really basic "what is AI/welcome to the first day of class" type stuff is up), so if you signed up, then sign in and hopefully the site will work for you.
I found out it's not even too late for the machine learning class so I signed up for it too. I'd read about Ng's work before and some of his projects, and I'm excited to take a class from him (not to mention Thrun and Norvig, of course).
Edit: I should add I was hesitant about taking both, which happen at the same time, but after seeing the first homework in the AI class, which was rather brief, I figured it would be manageable.
It is still possible to enrol in the machine learning class (I sucesfully registered in the advanced class).
Edit: I should add I was hesitant about taking both, which happen at the same time, but after seeing the first homework in the AI class, which was rather brief, I figured it would be manageable.
+1
+1 Back at ya. I'm beginning to think these courses are a little more basic than I was expecting. I had already been famliar with all the points Ng was making about linear regression, so I was skipping through most of it and putting him to 1.5x.
"We've had an absolutely massive surge of last-minute homework submissions" - how surprising!
"The massive crush of students trying to do homework at the last minute exceeded all of our expectations, and the expectations of our servers. To compensate those who have had difficulty accessing the site today we have moved the deadline by 24 hours."
That figures.
What's even more (pseudo-)surprising/depressing is that, at least according to this thread and subthread, only 1,000 people completed the first homework, out of the ~150k who signed up.
A commenter suggested that this is reasonable considering the wide net they cast for the course compared to the math pre-requisites, and the normal compsci class dropout rate (which increases for large classes).
The server seems to be in a bit of a clusterfuck, but unit 1 was interesting. Rot13: V npghnyyl zrffrq hc ba gur Puvarfr "fbhc/zvk" pbeerfcbaqrapr dhrfgvba, naq nz srryvat dhvgr flzcngurgvp sbe Tbbtyr Genafyngr evtug abj. :)
Yeah, I messed up with one of those too, but IIRC it was more a case of not noticing/missing one line.
(Also, part of the problem is that some of the symbols were partly covered by the buttons you could select, so..)
I got an email a week ago that said I'd be emailed again for relevant updates, but haven't been.
You can go to the site and sign in right now. Assuming the site behaves properly at the moment.
Cool, I was able to listen to the lectures yesterday evening and do the lecture quizzes. Perfect scores so far!
I wish they had transcribed versions of these lectures, they would go a lot quicker. Maybe I'll do it myself if I have time.
Tried it out, good stuff! Norvig speaks slower than I'm comfortable with. The transitions between videos and questions is a bit flaky though, and MySpeed might have something to do with it.
At just 150%, I found Andrew Ng (Machine Learning) somewhat hard to understand without guessing based on context. Does this software do a better job than the web player for the lectures?
I just tried, and the software player does seem to do a better job than the web player. It certainly sounds different. I don't have much trouble following Ng at 150% on the web player either, however, so it's hard for me to make a useful comparison.
I should say, though, that the machine learning videos don't seem to be loading fast enough on my connection for smooth sped-up streaming using the software. This makes the software pretty useless unless you pre-load the videos or are willing to put up with the frustration of intermittent buffering. I don't have this problem with the AI videos (or youtube videos in general).
OK here is my status if anybody is interested.
I am in the class for week 2. I read the book, did the classes-quizzes, did the homework, completed for week 1 (AI only.)
I found the reddit discussion page. I joined reddit after procrastinating that for a long time. The most interesting discussion there (to me) was the one on how many signed up, how many viewed the you tubes, how many are likely to finish the class.
I also joined another study group with 11 people off metafilter and we are all together there in the google+ circle and the google wave but there has not been any real activity there yet.
I have pages of questions and probably will trickle them into the reddit page mostly. The class "discussion" page where I am presuming Sebastian and Peter and/or their TA's are actually going to read the stuff is not up yet.
By the way, it's worth visiting the machine learning course site just to see the adorable robot-with-diploma mascot! :-)
There is a google group for the machine learning class (which I think still has open enrolment), with about 30 members but not much activity so far.
I can't speak for anyone else in the group, but since I'm enrolled in both classes I wouldn't mind discussion related to the AI class at all! :)
Thanks! How does ML compare to the AI class? I considered taking both, but I was worried overextending my leisure-time learning budget.
It comes off a bit more hands on with regards to implementation. For example there was a lot of talk about efficiency and the 5th segment is a quick tutorial for Octave, since programming assignments are also part of the class. The interface of the website is better currently (and for example the slides used in the presentations are available for download). The homework section which is just review for now can be taken almost as many times as you like (the variables of the problems change ala Khan Academy) and you get instant feedback. Ng comes off a bit better at presenting his material than Norvig but not better than Thrun (this is subjective obviously). I don't know quite what it is but there is something about the AI class which gives me the feeling that I will actually find it much more interesting than the Machine Learning class in about a month or so.
The first week of ML has proven much more demanding on my time than the AI class, but this is probably because I've watched most of the prerequisites for the AI class during the summer, but didn't touch any of the stuff that's been on the machine learning class website for two weeks or so until this weekend.
The on-line Machine Learning class seems more simplified compared to its Stanford version than the AI class is to its Stanford version (though the first week of the AI class was very easy, I understand the same lectures are used for the Stanford AI class). So far there aren't any indications you don't really need the AI class or something that gives equivalent knowledge before delving into the ML one.
I signed up for all three Stanford courses to see how they are before deciding which are worth the time investment. I have received a notification from both the Databank class and the Machine Learning class but not from the AI-class.
By the way: does LW have a learning group or something for this?
I just tried, and the software player does seem to do a better job than the web player. It certainly sounds different. I don't have much trouble following Ng at 150% on the web player either, however, so it's hard for me to make a useful comparison.
I should say, though, that the machine learning videos don't seem to be loading fast enough on my connection for smooth sped-up streaming using the software. This makes the software pretty useless unless you pre-load the videos or are willing to put up with the frustration of intermittent buffering. I don't have this problem with the AI videos (or youtube videos in general).