I have a bachelors degree in Computer Science and have been working as a programmer at a small financial company since I got out of school ~6 years ago. However, I've always had an interest in Artificial Intelligence and have been looking into potential Graduate Schools where I could get a degree in this subject. I don't know that I would actually do anything with it, but I read AI books for fun all the time so I might as well have a piece of paper proving I know something on the subject. Anyway, my particular question on this is about the University of Advancing Technology and their new degree in "Artificial Life". Here's a link to their very "flashy" site (pun intended). http://www.artificiallifedegree.com/ I realize this is not exactly Artificial Intelligence, as per their link on the front page explaining the differences, but the subject matter still greatly interests me. I very much like the concepts of Emergence, swarm intelligence, genetic algorithms, etc... Their example Ant video (programming each separate piece of the ant and then having the behavior emerge, versus just programming the whole ant itself) is exactly the kind of thing I would like to learn more about and investigate. So I did a search of this site for anything on UAT and I did find some matches, but no one mentioned the Artificial Life degree, possibly because it is too new? Does anyone have any knowledge on this degree? And is the UAT a well respected school in the field? Is it good for anything outside of Game Programming?

In my opinion, its not even good in the gaming industry. The whole subject is a bloated overhyped crappy field. If I was evaluating someone for a AI job and he came up with an "artificial life" degree, I would laugh at him, unless he had something better to show. Pretending to simulate biology does not make something worthy, but somehow the idea wont die.

But thats just my opinion!

I guess I'd start by asking: why do you want a post-graduate degree? I'm assuming you mean a PhD (I guess my comments still stand if you're going for a master's)

IMHO, if games is your goal, there's no point to getting a PhD unless you have something very specific you want to pursue.

The game industry doesn't actively recuit PhDs; sure it's something nice to have extra on your resume but the industry is literally starved for engineers at the moment. If you're competent at programming, know math and can demonstrate a passion for games you'll get hired. This isn't a NASA job or a Google job; a PhD isn't even remotely required.

There is precisely one person in my building of 100+ engineers that has a PhD. And his is in sub-atomic particle physics.

-me

Quote:Original post by Palidine
I guess I'd start by asking: why do you want a post-graduate degree? I'm assuming you mean a PhD (I guess my comments still stand if you're going for a master's)

I am only talking about a Master's here, at least for now. As for why? I don't have any immediate goals with it... I have always really liked AI and been interested in the subject. I read books all the time about it and I kind of figured as long as I'm studying this stuff on my own, I might as well do it under guidance and come out of it with a degree.

I guess I'd like to end up doing something in AI someday is all I know for now... what that is I just don't know.

Quote:
IMHO, if games is your goal, there's no point to getting a PhD unless you have something very specific you want to pursue.

The game industry doesn't actively recuit PhDs; sure it's something nice to have extra on your resume but the industry is literally starved for engineers at the moment. If you're competent at programming, know math and can demonstrate a passion for games you'll get hired. This isn't a NASA job or a Google job; a PhD isn't even remotely required.

There is precisely one person in my building of 100+ engineers that has a PhD. And his is in sub-atomic particle physics.

-me

I realize this is gamedev.net, so the replies are necessarily game development oriented, but I don't think I'm going to end up in the games industry. I do think it could be fun, writing the behaviors of the bad guys for a game (for example), but I don't have any direct plans to persue that path at the moment.

I'm mostly just generally interested in the field of AI and wanted some feedback from those of you out there already doing it.

Quote:Original post by Wyrd-One
I realize this is gamedev.net, so the replies are necessarily game development oriented, but I don't think I'm going to end up in the games industry. I do think it could be fun, writing the behaviors of the bad guys for a game (for example), but I don't have any direct plans to persue that path at the moment.

I'm mostly just generally interested in the field of AI and wanted some feedback from those of you out there already doing it.

Alright, fair enough. Just realize that most of this "A-Life" stuff is in fact very old AI disguised under a sugary cover. IMO for a master degree, having been through the process myself, the program is *nowhere* near as important as the advisor you'll get. Spend time to find an advisor that is flexible enough to let you work on the subjects you like, and motivated to make you obtain your diploma.

I'm studying robotic vision, and I know some of the Artificial Life concepts are applicable to robotics - my research group has close ties with the biology department. Things like swarm intelligence and modelling insects are useful when designing robotic control and vision systems. I'd also recommend going to a more standard research oriented university though.

From the website, it looks like "artificial life" is more a specialization than a degree. In any regard, it sounds a bit fishy. If the university needs to buy "artificiallifedegree.com" to advertise their degree, then that is probably not a good sign. Especially if you want to look outside of the games industry, consider a more research-oriented university with a reputation. Some good questions to ask UAT might be where graduates of their program go and what their job placement rate is.

Another program that might strongly interest you is Boston University's Cognitive and Neural Systems program. Many of their students come only for Masters degrees, and many have been out of school for a few years as well. BU's program is very applied, and their graduates go on to great jobs in the AI field, and even into the game industry!

I appreciate all the replies. The more I think about it the more I think it's probably best to go somewhere with a long history, and a well-respected record of turning out "real" students.

The fact that I'm not sure what I would even do with the degree at this point probably reenforces that stance. If I did decide to become a researcher in the field then I think a degree in AI from a well established University is much more likely to take me there than a potentially questionable "Artificial Life" degree from a mostly gaming oriented University.

So for those of you that have your AI degrees, did you have a good idea what you wanted to do in the field, and then went out and earned your degree?

I feel somewhat like a college student with an undeclared major. I know I'd like to do something in the realm of "A.I.", but inspiration hasn't struck me yet as to what that is. I'm trying to decide if it would be more valuable to start the degree and perhaps the ideas and concepts I'm introduced to will help guide me, or if I should figure out a definite direction I would take, and then go earn the degree with that goal in mind.

I also notice a subset of you feel that the more biologically-derived methods of AI are not really worth while. I find that discouraging. I've always thought (hoped?) there could be a lot of promise in such methodologies, a lot of potential that just hasn't been properly exploited yet. Perhaps for a certain class of problems something like a genetic algortihm is the best search method to find a solution? I have read that many have been quite successful at using an "ant algorithm" to find optimal paths within networks and similarly structured problems.