Hacker News new | comments | show | ask | jobs | submit login
Free Stanford AI Class is a “Beta” for a Commercial Launch? (dltj.org)
81 points by aheilbut 1860 days ago | hide | past | web | 33 comments | favorite

I could be wrong, but based on Know Labs own company description: "We're a Silicon Valley-based startup looking to change the future of education by making it more accessible and less expensive." I don't think the aim is to charge students for courses like ai-class, but rather quite the opposite.

It would look to me like the idea is to create a platform to allow universities to create their own version of ai-class. MIT gains a lot of positive press from OCW but it must cost them at least a little staff time and money to maintain and run it. I would guess there is definitely a market for a company to claim "Hey we're the guys behind Standford's famous AI class experiment, and for a price that saves you a lot of work and staff time you can have that same, well known, platform today"

Having worked in Universities for years, I know that at the institutional level 'buy-it' almost always trumps 'built-it'.

The AI Class official twitter account asked people whether they'd be willing to pay $2000 for a Stanford-level CS Master's degree taught in a similar format.

Do you have by any chance a link to that particular tweet? I have looked through the tweets since September and just couldn't find it. Would love to read what the feedback was on that suggestion.

It was from Sebastian Thrun's twitter account: http://twitter.com/#!/SebastianThrun/status/1123978353701150...

And more interestingly he follows up a few weeks later claiming that standford could actually offer the masters for free! http://twitter.com/#!/SebastianThrun/status/1171319537146429...

Free master's in CS from standford is a pretty mind blowing idea. Would be really curious to see what would happen to higher ed if you completely removed/minimized the cost factor.

Thrun had some interesting things to say about his vision for the future of higher education:

Thrun’s ultimate mission is a virtual university in which the best professors broadcast their lectures to tens of thousands of students. Testing, peer interaction and grading would happen online; a cadre of teaching assistants would provide some human supervision; and the price would be within reach of almost anyone. “Literally, we can probably get the same quality of education I teach in class for about 1 to 2 percent of the cost,” Thrun told me.

The traditional university, in his view, serves a fortunate few, inefficiently, with a business model built on exclusivity. “I’m not at all against the on-campus experience,” he said. “I love it. It’s great. It has a lot of things which cannot be replaced by anything online. But it’s also insanely uneconomical.”

Thrun acknowledges that there are still serious quality-control problems to be licked. How do you keep an invisible student from cheating? How do you even know who is sitting at that remote keyboard? Will the education really be as compelling — and will it last? Thrun believes there are technological answers to all of these questions, some of them being worked out already by other online frontiersmen.

“If we can solve this,” he said, “I think it will disrupt all of higher education.”


It is also interesting that he phrases it never as a Stanford degree but as a degree "of Stanford quality". It is going to be interesting to see how they will leverage the Stanford brand without discounting it.

I'm partaking in both the AI and ML classes and I find the ML experience much more user-friendly and polished than the AI class. For example, being able to speed up the lecture videos is brilliant. I find the AI class video questions a bit frustrating to use. IMHO, Know Labs has a ways to go.

Agreed. The ML one is much more organized and the design and experience are both incredibly better. It has more of a classroom and course feeling to it, rather than the AI class's here's some video lectures and quizzes, now do the tests.

The ML class also has a full-fledged Q&A forum (which has LaTeX markup support), which, to me, is a necessary component. The about page[0] lists Stanford students as developers (the first one is an undergraduate sophomore named Frank Chen), and the team is different than the guys behind Know Labs (from a quick search.)

Andrew Ng is an excellent teacher. Seriously, the ML class is the most fun I've had learning in a college course environment. That speaks volumes.

[0]: http://www.ml-class.org/course/aboutus/index

It's kind of funny how a summer project of 3 students lets the effort of a backed SV startup look like a project from a 13 year old child.

I mean what are "knowlabs" even contributing here? The Videos are hosted on Youtube and recorded using... PAPER???

Ok you might say: but there are quizzes!! Those quizzes are the least userfriendly quizzed I've seen in my life. ML & DB-class are way ahead on this front.

for what it's worth, I'm taking Stanford's version of the ML class that Prof. Ng is offering, and aside from being a couple of weeks and ahead and having to do a final project, it is pretty much the same offering as that available the outside world. I think this could explain the level of quality of the website, or at the very least it having more of a course feeling to it.

I share your opinion.

I wish I could spend the whole week to dive into AI, IMHO without previous knowledge a regular person can't catch up with course progress. Unfortunately I can't leave everything behind to study the deeps of AI and in this case the ML does a great job, Professor NG presets lectures and a few quizes to see if you understand that sensible topic, then you can apply those concepts with programming exercices and homework.

I'm almost giving up about my dream that is attend a course in Stanford. I'm a 28 college dropout who had to start working to help family. I don't live in US, I'm married, I have a great job (that I believe is far more interesting than most of you guys). But I have this itch for Academics. *

I'm glad I've been playing around with both ML and AI classes but I think that Stanford should stick with Professor NG didactics.

ps: I know that my statement sounds like the "occupies" but is hard to see dreams not becoming reality. I feel sorry for those who dropout Stanford/MIT to create a landing page startups.

Seconded. I've had to drop AI class because the lack of downloadable videos means I can't watch them when I like - out internet speed is appalling when I'm free to do things like this. Still going with db-class though.

It does make me wonder why ai-class went its own way with the website and delivery methods. especially as db-class and ml-class worked closely together on it.

the lack of downloadable videos means I can't watch them when I like

Take a look at this: http://www.reddit.com/r/aiclass/comments/lqm0e/an_easy_comma...

not sure if it is of interest to you anymore but there are a couple of scripts around the internet to scrape ai-class.org & youtube to get you offline videos.

FWIW, I'd _love_ if you could download a single file "something" (mhtml, swf, pdf or heck even one of those new wolfram document format) including both all video lectures for a single db/ml units, subtitles and quizzes.

Yes the db and ml class websites are outstanding. The video and in class exercises are very smooth, but the best part are the exercises in both classes. The db teacher said some students threw all of it together over the summer. Very impressed.

Interestingly it appears that the ML-class (http://www.ml-class.org) and DB-class (http://www.db-class.org/) are not associated with Know Labs. But I might be wrong. I initially thought all three of them was a part of a Stanford effort to push their own name even further (perhaps in response to the online MIT lecture videos that have been quite popular in recent years).

My daughter is now looking at colleges. These courses both from MIT and Stanford give me, the one who will be paying, a sense of what my 70K per year will be buying. I think this is great marketing for less cost than those mass mailings all the schools seem to be cramming into our mailbox. I just wish all these tier 1 schools would offer something online, it would make judging the quality of what I will soon be buying a lot easier.

I do not want exactly to complain since the class is currently free and I am enjoying it quite a bit, but there is a lot of issues. The integration between the Youtube video and the quizz is poor, no official forum for the student, no lecture notes. Everything as a very beta feels to it.

On the contrary the platform for the ML class is of very high quality and the content more polished. I especially like the possibility to speed up the video lecture, having programming assignment, and how easy it is to submit them.

Some of the best things in life are commercial and still have a phenomenal impact on humanity...

I understand why people might be wary, like if the Khan academy suddenly opened up premium memberships, but this course was never designated as a non-commercial concept (though it was very much billed as a 'beta')

A bit of amateur Google-sleuthing doesn't really amount to a groundbreaking piece of news, in my opinion.

This isn't really news. Ai-class.com links directly to Know Labs website.

The way it was hyped when initially announced made it sound very much like it was Stanford running this, and that was my presumption.

I happened to be curious after a conversation with a friend about the course, and when I went to have a look at one of the homework assignments, realized that they aren't available to those who haven't signed up for the class. That's just fine, and I respect that Know Labs and the instructors can do whatever they want (and it's fantastic that they are making a lot of the materials available). However, it struck me that this was marketed as being about open education when the reality is that it has the potential of being much more locked down than MIT OCW or the Stanford courses that do have lectures, complete syllabi and assignments freely available on the web.

The difference is that this is a class that you participate in real-time. Each week you cover the same material as everyone else in the class. The alternatives that you mentioned are all self-paced. Each has it's merits, but being able to form study groups for the class is helpful, and not likely to happen with some of the MIT OCW materials.

And yet you can still sign up for the ml-class.org and db-class.org, all assignments that you missed and will not be doing "in real time" have 50% penalty, and you are free join in at any time. For ai-class.com, however, the decision was made to close it up the moment the classes started.

This makes a lot of sense. I was working through a couple of the assignments last night when I realized that a lot of the questions seemed to test reading comprehension rather than proper understanding of the material. I had a while back used Prof. Ng's video lectures and notes (available on his site) as a supplement to my own machine learning class and those are designed to both stretch your mind and test your understanding. I was kind of hoping that that level of quality was being shared with the rest of the world but that doesn't seem to be the case :(

for what it's worth, I'm taking Stanford's version of the ML class that Prof. Ng is offering, and aside from being a couple of weeks and ahead and having to do a final project, it is pretty much the same offering as that available the outside world.

I read that some things were dropped from the online version. Specifically, reinforcement learning. Is that still part of your syllabus?

Edit: Here's the link. Not sure if you can see it without being logged in. http://www.ml-class.org/course/qna/view?id=58

That's exactly the discussion I had that led me to go and want to check out the site. Apparently the pace and nature of the (AI class) homeworks are quite different in the actual Stanford course.

Guess it has a lot to do with having to be corrected automatically rather than a professor going through your process on questions with individual correction.

That might actually be an interesting AI research area, making computers better at marking based on process the student used rather than just the final answer. Wonder if much has been done in that area as it will become more important as more learning goes online and people want more than multiple choice/ final numerical answer type stuff.

They can probably infer it based on the distribution of answers.

They can probably infer the mistakes people made by looking at the really common but incorrect answers. I'm guessing that lots of people get the same wrong answers as each other.

Yes, at is says on the http://www.knowlabs.com/ website.

I don't know if Knowlabs is testing it to commercially launch it. But i saw a tweet from Sebastian Thrun's twitter account a month or so back asking 'if you're provided with Stanford quality education with a tuition fees of $2000 , will you take up the course' i dont know what replies he got.. It intrigues me.

I wonder if this experiment will just remain an experiment because politically Stanford is not ready for open free education. I remember reading an article recently in NY Times that highlighted how Stanford had two opposite camps of opinions. They are bidding to open an engineering campus in NY, which expands their formal education at the same time as starting these free courses, which is all about informal distributed learning. From Stanford's president, it didn't sound like something he was keen on expanding... a classic disruption in the education industry. Reminds me of newspapers.


Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact