Data for 2009 + 2010 March Madness. Can you use machine learning predict the tournament?

llimllib

If you're interested, you may want Ken Pomeroy's raw data, available in a csv here; I have last year's (pre-tournament) data archived here in HTML format, not csv (why? I dunno.)

Anyway, using his stats you should be able to get a baseline probability for any team beating any other using the pythagorean fomula.

llimllib

hilarious... I used to listen to 104.1 incessantly too, though I remember it as being poppier than you do. (i.e. when I became 16, had a car and started listening to "cool" music, I got much too "cool" for that radio station)

llimllib

That's why I included the asterisk.

What you said is obviously true, but I think it's interesting nonetheless.

llimllib

squarepusher, aphex twin, four tet, amon tobin, dj shadow, jaga jazzist, burial, prefuse 73, boards of canada, nosaj thing, Autechre, and holy fuck

But I'm much too lazy to youtube search for him :)

llimllib

So yes, compared to what the same person could be making elsewhere, they do take a pay cut.

Right, I should have made it clearer that they can still make bank by any reasonable standard. OP should consider that the salary for a very good programmer in any field is a very livable wage and he should make his decisions based more on his own happiness than on what field pays more than what other.

llimllib

3) Very good - if you can get in, and to the top.

It's been my impression that the glut of people who want to be in video games leads to depressed salaries for video game programmers. cf Oliver Reeves:

I've certainly given it a shot in the past. I took a substantial pay-cut to work for Electronic Arts back in 2005 so that I could work on something that I really wanted to work on. I thoroughly enjoyed it. As far as the technology and the challenges are concerned, it's the most fun I've ever had at work. Unfortunately, the experience I had there led me to believe that the lifestyle isn't sustainable (at least not for me and my family).

So my answers would be:

1) yes

2) no, I've never been very interested in video games. I love comp theory and Sipser (seriously it's one of the best textbooks ever, so concise and to the point. I've reread it for fun), so that abstruse goose comic doesn't apply to me :)

3) not as good.

edit to answer the title question:

CS led me to a job where I get to think, hard, every day. I get to do a job where I get to read academic papers, write and patch open source software, do basically zero shit/adminstrative/paper work, and play FIFA with my boss every day at lunch. Neato!

llimllib

"A demonstrator throws a piece of marble to policeman..."

What a nice guy!

llimllib

and the software he references is available here.

llimllib

There's a documentary about Bill Withers coming out soon that looks awesome.

llimllib

Thanks for the hard work... the site works great for me.

I made the cover and labels for my CD today, it goes out the door tomorrow. Sweet!

llimllib

I went to a small liberal arts school not known for its program, and I got an excellent education in large part by working with the profs. Ask to help them with research or to see if they can get you credits for independent research.

I did both, and benefitted greatly.

Also, don't just complete your assignments, dominate them. Extend them, if there's 2 options pick 1, do both. If it says you don't have to do all of X to complete this, do all of X. Complete the assignment in 2 different languages, one you're comfortable with and one you're not.

Ask professors if you can TA the freshman lab, and see if there's a math or computer science help room you can work in; both are ways to earn a little drinking cash while getting to make sure you're sharp on what you should have mastered.

llimllib

ooh network theory I forgot about that. I needed to know CS so I could read papers like these and improve our model of the internet.

llimllib

I'm a programmer for a "cyber security" (god I hate that phrase, but there you go) slightly post-startup small business.

edit to be more complete: What I do, lately, is write code to archive somewhat silly amounts of data that we gather on the internet every day, and try to do it without using up a billion gigs of storage and in such a way that it can be searched in a reasonable amount of time. So because I'm working with lots of data, it's crucial to know the complexity of the algorithms I'm using.

Furthermore, knowledge of what algorithms are out there means I can choose well from the beginning.

llimllib

yes, every day. I couldn't do my job if I didn't know data structures, algorithms and algorithm analysis.

llimllib

If you'd like to understand it better, here's your assignment:

Write your original question out in two different ways (in english, math, or any combination of the two) such that the solution I gave is clearly the answer.

Sometimes restating things in different ways will make the answer jump out at you such that you'll wonder how you ever wondered what it was; I think this might be one of those cases.

edit: besides english or math, you could also do a "visual" proof of why this is the case.

llimllib

Add them together and divide by two?

That is, if f(x) = x+1 and g(x) = x^2, then the average function avg(x) = (x + 1 + x^2) / 2

edit: The question mark is there because you already seem to understand the answer (average the y values at every x) and I'm wondering if I'm the one who's missing something.

llimllib

no worries, I'm sure there's lots of stuff to keep track of. You're doing awesome work, I'm excited both to send and receive mine.

llimllib

TROPHY CASE

you'll need to login or register to do that

create a new account

login