The new forums will be named Coin Return (based on the most recent vote)! You can check on the status and timeline of the transition to the new forums here.
Please vote in the Forum Structure Poll. Polling will close at 2PM EST on January 21, 2025.
I'm going to be a senior next year, and I'm looking at colleges now. I've already checked out UAT and Full Sail. I REALLY want to go into Aritifical Intelligence (NOT Artificial Life, which is what UAT offers. That means my main college is Full Sail), thanks to my visions of grandeur. I made a 26 the first time through the ACT, and I'm taking it again on June 9th. My question is, are there any other notable colleges/universities I should be keeping my eyes peeled for that are a landmark in the field of Artificial Intelligence? Please and thank you.
Honestly, it really doesn't matter for undergrad. Just pick a good uni and get a degree in EE, CompE, Math, CS, or something similar. You can do beginning AI in any of these. You WILL NOT be doing real AI work until grad school. You may be able to start getting involved in undergrad research for AI, but that will most likely be limited work for you.
You also have to remember there are divisions of AI too.
That said there are plenty of good state and private schools in the US with great science and engineering schools.
Purdue, UIUC, UMichigan, UW-Madison, UC-Berkley, MIT(duh), etc.
It's like wanting to become a car mechanic and then only being qualified for tyres. You had better have a lot of jobs that need only tyre experience!
Take a Computer Science major, with an AI minor if offered, or just AI type classes. I say this because it will help you become much more rounded out as a student, it will give you the tools you need to perform AI work, and you may find that you actually don't like AI at all, and now have the skills to do something else.
I have been researching studying AI as my PhD thesis. All the schools you associate with good Comp Sci also do good AI work: USC, Stanford, Berkeley, those types. Choose a university that has an AI or Robotics research lab, then you are more likely to find professors who have a deep AI interest and offer better classes as a result.
You're in a funny situation, actually. AI is one of the pillars of classical (e.g., 1970s-era) computer science. 1970s-era computer science still forms the backbones for many, if not most of the CS curricula offered at good universities around the world. Therefore, a good CS curriculum will undoubtedly include some measure of AI along with other pillars like algorithms, compiler design, and so on.
Now, there are many who argue that given AI's relative failure to live up to its initial promises, it should be demoted somewhat out of the mainstream, with an increased focus on more practical areas like software engineering, team management, and so on. However, luckily for you, this idea hasn't caught too much fire yet.
Most of the old-guard CS schools (MIT, Berkeley, Stanford) still have serious AI heritage and still do it for real. Most good CS curricula include at least a concentration in AI or some similar area (data mining or 'knowledge-based' systems come to mind). AI isn't really something you can or should "major" in anyway. As others have suggested, get a broad-spectrum CS degree with an AI concentration, and then if it really floats your boat pursue it in grad school.
Can I give you a pretentious warning? AI = math. Hard math.
I have a good friend/classmate who interned for EA and spent the better part of a summer writing code for the reactions that soldiers have to grenades and enemy fire for Call of Duty 4. He's a grad student at Stanford in CS, with a specialization in AI. He said that it was the most intense math and CS that he's ever done, because not only do you have to know the theory in and out, but you have to be able to optimize the shit out of it, which means lots of math tricks.
Just make sure that's what you like. I'm doing CS, but I like the engineering side of it much more. Start out broad, then narrow down later.
I have a good friend/classmate who interned for EA and spent the better part of a summer writing code for the reactions that soldiers have to grenades and enemy fire for Call of Duty 4. He's a grad student at Stanford in CS, with a specialization in AI. He said that it was the most intense math and CS that he's ever done, because not only do you have to know the theory in and out, but you have to be able to optimize the shit out of it, which means lots of math tricks.
This is partly true, but it very much depends on how theroetical you go and what the professors expect of you; Stanford has some of the most well-known AI researchers in the world.
My Math is absolutely atrocious, and I am looking to research Agents (autonomous self-centred AI programs), and I have been getting along fine with the mathematics. I think this is because Agents are really a sub-set of AI research, and most of that hard math is down that end of the spectrum, agents are really only involved in decision making.
That said, I am shitting bricks at the idea that I won't be able to handle the math when it counts
For MIT and whatnots, no offense man, but a 26 is somewhat on the low end for really high end schools like that. Your first step is to aim for a higher ACT. I knew a guy that was a valedictorian, had tons of extracirrulars, amazing essay, and got a 35 on the ACT and still didn't get into MIT.
Don't bother with something like MIT, it's not worth it for undergrad. Focus on general quality and, depending on how large your trust fund is, cheapness.
So I would reccomend applying to Canadian schools, first of all. I'm attending McGill University, which at about $15,000 a year is almost the same price as my state school. It's also the most expensive one in Canada, to my knowledge. Going to University of Toronto gives you a big-city education at tiny-school prices.
Second, you're going to change your major, most likely. Going to a really specialized school is a bit of a gamble, especially as you cannot possibly have had a ton of experience in AI as of yet. Instead of something like Full Sail, I would recommend something more well-rounded, that still had the potential to teach what interests you right now.
In my case, I was interested in computer science, but not enough for a major. I developed an interest in Pyschology and Cognitive Science. Doing a Major in Psych and a minor in CogSci and CS has been interesting so far, and pretty rewarding. I've noticed that most people interested in things relating to computer science tend to focus exclusively on higher-level math, which in my opinion leads to the "specialized mechanic" thing Lewisham mentioned. Approaching it from the perspective of another major can help round out your interests, as well as give you some interdisciplinary perspective. Plus, my Psych/CS profs who are interested in AI love that I am doing a combination.
Though, yeah, the math frightens me a bit. Though if you're interested in AI I'm going to go out on a limb and assume you're at least reasonably capable at math. In my case, the profs I've spoken to in Pysch seem to be focusing on Neural Networks, whereas the CS profs are more focused on expert systems and slightly less "pure" AI. The CS stuff tends to require more math work, the Pysch stuff more theorizing.
Basically, if your interest in AI is specific, abiding, and focused mostly on math and application, go for a school that has your specific program. If it's more focused on theory, research, and design, I would go for a more generalized program, in which case you can probably focus more on price.
Seriously though, apply to Canadian schools, the process is pretty painless, and odds are every one will cost less than or as much as your state school.
Oh! Also. I'm trying to get back into RoboCode. You might find it awesome as well.
Posts
You also have to remember there are divisions of AI too.
That said there are plenty of good state and private schools in the US with great science and engineering schools.
Purdue, UIUC, UMichigan, UW-Madison, UC-Berkley, MIT(duh), etc.
It's like wanting to become a car mechanic and then only being qualified for tyres. You had better have a lot of jobs that need only tyre experience!
Take a Computer Science major, with an AI minor if offered, or just AI type classes. I say this because it will help you become much more rounded out as a student, it will give you the tools you need to perform AI work, and you may find that you actually don't like AI at all, and now have the skills to do something else.
I have been researching studying AI as my PhD thesis. All the schools you associate with good Comp Sci also do good AI work: USC, Stanford, Berkeley, those types. Choose a university that has an AI or Robotics research lab, then you are more likely to find professors who have a deep AI interest and offer better classes as a result.
Luckily I majored in CS and now I work at software engineering at EMC on data storage
So Yeah, what Lewisham said
Now, there are many who argue that given AI's relative failure to live up to its initial promises, it should be demoted somewhat out of the mainstream, with an increased focus on more practical areas like software engineering, team management, and so on. However, luckily for you, this idea hasn't caught too much fire yet.
Most of the old-guard CS schools (MIT, Berkeley, Stanford) still have serious AI heritage and still do it for real. Most good CS curricula include at least a concentration in AI or some similar area (data mining or 'knowledge-based' systems come to mind). AI isn't really something you can or should "major" in anyway. As others have suggested, get a broad-spectrum CS degree with an AI concentration, and then if it really floats your boat pursue it in grad school.
I have a good friend/classmate who interned for EA and spent the better part of a summer writing code for the reactions that soldiers have to grenades and enemy fire for Call of Duty 4. He's a grad student at Stanford in CS, with a specialization in AI. He said that it was the most intense math and CS that he's ever done, because not only do you have to know the theory in and out, but you have to be able to optimize the shit out of it, which means lots of math tricks.
Just make sure that's what you like. I'm doing CS, but I like the engineering side of it much more. Start out broad, then narrow down later.
This is partly true, but it very much depends on how theroetical you go and what the professors expect of you; Stanford has some of the most well-known AI researchers in the world.
My Math is absolutely atrocious, and I am looking to research Agents (autonomous self-centred AI programs), and I have been getting along fine with the mathematics. I think this is because Agents are really a sub-set of AI research, and most of that hard math is down that end of the spectrum, agents are really only involved in decision making.
That said, I am shitting bricks at the idea that I won't be able to handle the math when it counts
So I would reccomend applying to Canadian schools, first of all. I'm attending McGill University, which at about $15,000 a year is almost the same price as my state school. It's also the most expensive one in Canada, to my knowledge. Going to University of Toronto gives you a big-city education at tiny-school prices.
Second, you're going to change your major, most likely. Going to a really specialized school is a bit of a gamble, especially as you cannot possibly have had a ton of experience in AI as of yet. Instead of something like Full Sail, I would recommend something more well-rounded, that still had the potential to teach what interests you right now.
In my case, I was interested in computer science, but not enough for a major. I developed an interest in Pyschology and Cognitive Science. Doing a Major in Psych and a minor in CogSci and CS has been interesting so far, and pretty rewarding. I've noticed that most people interested in things relating to computer science tend to focus exclusively on higher-level math, which in my opinion leads to the "specialized mechanic" thing Lewisham mentioned. Approaching it from the perspective of another major can help round out your interests, as well as give you some interdisciplinary perspective. Plus, my Psych/CS profs who are interested in AI love that I am doing a combination.
Though, yeah, the math frightens me a bit. Though if you're interested in AI I'm going to go out on a limb and assume you're at least reasonably capable at math. In my case, the profs I've spoken to in Pysch seem to be focusing on Neural Networks, whereas the CS profs are more focused on expert systems and slightly less "pure" AI. The CS stuff tends to require more math work, the Pysch stuff more theorizing.
Basically, if your interest in AI is specific, abiding, and focused mostly on math and application, go for a school that has your specific program. If it's more focused on theory, research, and design, I would go for a more generalized program, in which case you can probably focus more on price.
Seriously though, apply to Canadian schools, the process is pretty painless, and odds are every one will cost less than or as much as your state school.
Oh! Also. I'm trying to get back into RoboCode. You might find it awesome as well.
I remember that as part of a class I was given in my first year... I should totally setup a GameOn for it in G&T...