Calc 3: worth it?

[DaniMetroplex]

I should finish my BS in Computer Science (possible math minor) in December 2010. I'll start by giving my current stats:

School: University of North Texas

GPA: 3.8

GRE: 760 quant (Do schools even care about the other scores if I'm US-born and can put a coherent sentence together?)

Interned for 2 summers with Fortune 500 company

No research experience but looking for ways to remedy that

My actual question: I've had two semesters of calculus and will take linear algebra before I graduate. Would Calc 3 be of any help to a possible grad school application? Specifically, could it make enough of a difference to be worth paying out of pocket?

Thanks,

DaniMetroplex

[belowthree]

Doubt it would make much of a difference unless you were looking at a math-based subfield, which there are several in CS.

So what type of work do you plan to do? Your qualifications do look a little slim for a PhD program perhaps... you might want to spin through a Masters first and grab some research experience.

[DaniMetroplex]

First, thanks for your reply, belowthree!

I would like to study Artificial Intelligence. Does everyone and their brother say that? From what I've read, it seems like it can be quite math-intensive, but perhaps statistics and probability courses might be more useful than more calculus? If I am wrong, please correct me.

Since you brought up getting a Master's, I'll go ahead and ask a question about that. There's no doubt I would be welcome in the Master's program here at UNT, and I could probably get a teaching assistantship after my first semester. If I were to get some decent research experience (I'd pretty much have to in order to complete a thesis, yes?), would that at least somewhat overcome the lack of school prestige? Or should I try to get into at least a slightly more prestigious program even for a Master's? Understand that I have no delusions of MIT or Berkeley, but I don't want to hear, "Where?" or "I'm not familiar with that department" for wherever I get my future PhD.

Basically, if I bust my hump on a Master's at UNT and get some research experience, will that be enough for at least a decent PhD program?

DaniMetroplex

[belowthree]

I would like to study Artificial Intelligence. Does everyone and their brother say that?

No, I haven't noticed any particular subfield that people universally gravitate too... everyone seems to develop their own likings.

From what I've read, it seems like it can be quite math-intensive, but perhaps statistics and probability courses might be more useful than more calculus? If I am wrong, please correct me.

I know there's a lot of math in that part of the field, but outside of that I'm pretty clueless with what the AI folks are up to.

Since you brought up getting a Master's, I'll go ahead and ask a question about that. There's no doubt I would be welcome in the Master's program here at UNT, and I could probably get a teaching assistantship after my first semester.

That sounds like a decent deal. Do you know the professors well enough where you'd have a route to a research assistantship a semester or two later?

If I were to get some decent research experience (I'd pretty much have to in order to complete a thesis, yes?), would that at least somewhat overcome the lack of school prestige?

Yes, you'd want to take the research-based masters route if your program offers a choice. Your goal would be to do good world-class (seriously, world-class) research that shows admission committees you're ready and capable to step in on day one and impress the hell out of them.

Or should I try to get into at least a slightly more prestigious program even for a Master's? Understand that I have no delusions of MIT or Berkeley, but I don't want to hear, "Where?" or "I'm not familiar with that department" for wherever I get my future PhD.

If you think you can stay were you are and be funded and do good research there, then staying seems like a reasonable idea. If you do good work you'll be able to move someone people have heard of for your PhD. You may not have a particularly need for MIT or Berkeley, but you should shoot for that level of work anyways and then allow yourself to be mildly disappointed if they end up rejecting you. (At which point you can accept some other wonderful school's offer.)

Basically, if I bust my hump on a Master's at UNT and get some research experience, will that be enough for at least a decent PhD program?

Yes... if you can do good work there, it will be recognized. The better schools are better because it tends to be easier to do good work there (and they tend to demand it more often) not because people going there are inheriently better. If you feel like you can do good work at UNT then that work will be recognized more or less fairly. (I mean, yes, there will be a slight coloring that the work was from UNT, but sometimes that coloring can work in your favor, i.e. even if faculty member thinks badly of your school they may still go "wow this is impressive work for a student from there".)

[DaniMetroplex]

Sorry if this sounds dense, but what distinguishes good or world-class research from bad or ordinary research?

Thanks for your patience,

DaniMetroplex

[belowthree]

There's no simple way to answer that. You get a feel for it after awhile.

Until then, you can use where the research is published as a rough guide. For people not familiar with your sub-field, this is likely what they'll have to do anyways. If something's published at a top-notch world-class conference that generally means that it had to beat out at least some other high quality research to get published. This isn't always true and sometimes things just happen for odd reasons. But it's a close approximation and usually what adcoms rely on.

If it's published in a good place, it'll be viewed as good research. Unfortunately this often means that good research published in a bad venue gets overlooked. Although if someone in your field actually reads the paper (likely to some extent if your application gets beyond a certain point) they might be able to tell, but this will only happen if they determine you're promising and if you publish in a junk venue you might not ever get that far.

[DaniMetroplex]

I was going to ask how one determines what the good conferences are, but a few quick searches revealed some very useful information on that issue. I suppose the next order of business is to comb through the proceedings of said conferences (and maybe a few of the junk ones for contrast).

Regarding the original question, I think I'm going to hold off on Calc 3 for now. If I do end up taking extra math, it will probably be in the differential equations direction. And there's always OpenCourseWare.

Thanks so much for your help!

DaniMetroplex