Stanford "Intro to AI" course to be offered online for free | Hacker News Reader

Stanford "Intro to AI" course to be offered online for free

spectrum.ieee.org

150 pointseguizzo25 comments

Previously (with discussion): http://news.ycombinator.com/item?id=2824780

I can't wait to see how well this works out. If the results are good it could put another nail in the coffin for classroom only education.

Although I foresee a lot of those that register for the free online course dropping out. Most people won't put a high enough value on a free class, especially if they're not getting credit.

> Most people won't put a high enough value on a free class, especially if they're not getting credit.

Both Sebastian Thrun and Peter Norvig are Very Important Persons in Google (Google Fellow / Google Director). This is basically a free diploma to be put in the CV for any Google wanna-be engineer, if they can make it. I doubt even credit would be more important than it.

Well, even if you're not a Google wanna-be it's pretty cool. I signed up (for an update when registration is open), but I'm afraid I won't have the time to properly devote to it as well. If it's free of charge learning there is a tendency to let other things get in the way vs. prioritizing it because you paid full tuition price. At least that was my experience in college with the trust fund kids vs. the people that were working three jobs to scrape tuition together. There may be some selection bias there though.

I agree with your assesment that many (most?) will sign up without truly giving it the effort required. I wish it was $20 - some token amount. To extrapolate metrics on 'paid' users vs free users on most freemium apps, the vast majority of people won't even open their wallet for a dollar. If you aren't serious enough to pay $20 for the class, you aren't serious enough to buy the book and do the homework.

I'm really glad it's free. Where I live, $20 pays a month of lunch or public transportation. Even if it were $1 or $0.01 it would exclude people without international credit cards.

What's the 6th word on the 3rd paragraph of page 89?

This will not even come close to threatening instructor-led education. There are a very small number of courses that can be automated in this way.

I also think that the instructors are going to be surprised when the 10,000 students come up with more than 25 interesting questions.

I wonder what happens for the students who need questions answered that aren't deemed sufficiently interesting or unique?

Looks like I have two months to brush up on my probability and linear algebra. Could anyone who's read AIMA or taken the Stanford course suggest which topics I should (de)emphasize?

has anyone started reading the AIMA? maybe we should set up a study group?

I am interested. What do you think it is permitted regarding collaboration ?

Anyone who's interested in joining a study group, I've added a Google Doc spreadsheet: http://goo.gl/cbzUL.

Based on the lack of comments here it's hard to tell what the overall interest level is for a study group, but personally I've found them to be quite rewarding.

As the course's start date approaches we'll see how many people have signed up and take it from there.

It might turn out a better alternative to doing this in an ad hoc manner to use one of the online study groups (like OpenStudy, for instance). Personally, I haven't any experience with them yet, but I'm sure there are others who have and can shed some insight.

What is the big deal? If you really wanted to learn AI there are plenty of books and free resources online to do it. Why wait to learn until somebody offers a free course online?

Peter Norvig.

Awesome, thanks!

Some people seem to be very excited by this. You don't have to wait until October. Andrew Ng's course on Machine Learning@Stanford is available for free on iTunes University (it also seems to be on YouTube). Do note that ML is a subset of AI. Also, my experience has been that the assignments and exams constitute a large fraction of the learning in a course. For that, you'll have to wait until October :)

It might be easier to take 229 (Machine Learning) after 221 (this class), since 221 is intended to be a lead-in. and less mathematically rigorous. It's still definitely worth a look but may be hard to understand without a background in algorithm design.

this has been posted like ten times

It is the first time that I have seen it though -- and I check HN far too frequently.

This is total awesomeness for the really motivated to learn. I don't think I'll have the time... although, maybe this is a good enough reason to quit my job. This is certainly an indication of what the future will bring for education. I hope other universities/professors do the same...

Do you know if you'll need the 3rd edition of AIMA for the class? I have the 2nd edition and I would rather not buy the book again.

"Without having to live in a dorm" ... I thought living in a dorm was the best part of college.

Then you didn't live in a building full of 200+ proto-adults, many of whom lacked bsaic cleaning skills, long enough... or maybe I just lived in one too long :)

Always wanted to do an AI course, would be signing up right away ...

This is awesome !!! :)

Reply on news.ycombinator.com