Math PhD Chances: UIUC and CMU

[Bourbaki1123]

I have my profile typed up https://www.physicsforums.com/showthread.php?t=520174", but that thread seems to have become mostly about stats, my interest here is in my chances at a couple of institutions that feel like reaches to me, but in areas that I have a lot of background in.

So, can anyone give me an idea of my chances at Carnegie Mellon Math with first choice of Pure and Applied Logic and UIUC, where I would like to study algebraic geometry (particularly with Iwan Duursma since (a) he is doing stuff I consider interesting across the board and (b) some of his interests are things I've done research in already (algebraic cryptology)).

I am considering contacting a couple of professors at each school who do work I find interesting, would this be recommended? Should I inquire as to the likelihood that they won't have too many students they will be advising should I be accepted to the program, or is that gauche?

 

[Vanadium 50]

What do you care what your chances are? The only thing we can say with certainty is "if you don't apply, you won't get in". If you decide not to apply because of the guesses of some random people on the internet...

 

[streeters]

What do you care what your chances are? The only thing we can say with certainty is "if you don't apply, you won't get in". If you decide not to apply because of the guesses of some random people on the internet...

Yeah, just apply. It doesn't matter what anyone thinks. Except for the admissions board, that is.

 

[Bourbaki1123]

I'm going to apply either way; in no way is this going to affect whether I apply, it was not my intention to imply such. It may affect how wide a net I cast in order to boost my chances; there are too many schools for me to apply to that I would be happy attending, so I'm trying to gauge how strong my profile is relative to other applicants in order to figure out what the best strategy for applying would be.

That said I was hoping that someone might have come across relevant information, such as applicant profiles and where they were accepted that I might have missed or might have had a similar profile and could share where they were admitted.

I've been checking out grad cafe and the mathematicsgre.com site to get some ideas, but additional information would be appreciated.

Also, do either of you have any input as to whether it is appropriate to contact professors at the places you will be applying if their work is of interest to you?

 

[Vanadium 50]

You should apply to as many schools as you would be happy attending. I would expect that to be 5-10.

 

[Bourbaki1123]

It's significantly more than that; I have multiple fields of interest and would be happy attending graduate school for any of them.

My main areas of interest are computational cognitive science, theoretical neuroscience, algebraic geometry, proof theory (I'm presently interested in Girard's linear logic and game semantics more generally), statistical inference/machine learning and cognitive/evolutionary psychology.

Statistics/machine learning and the various cognitive/neuroscience disciplines are areas that I've mostly studied on my own over the past year so compared to mathematical logic or algebra my knowledge of them is still fairly shallow; however there are an awful lot of people who apply to schools in the cognitive fields (and to some extent statistics, as I understanding, since math majors tend to jump into stats phds) who have little to no official background due to the lack of undergraduate majors in those subjects and their interdisciplinary nature.

I've mostly figured out where I plan to apply for everywhere except for math programs, and I've got easily a dozen places that are of interest to me. I like Rensellaer because of their diversity of applications, I like UIUC because of all of their algebra people; I like CMU because of their Pure and Applied logic track; I like Ohio State because they have people working in reverse mathematics (which is incredibly fascinating to me, I've read a decent amount of Simpson's Subsystems of Second Order Arithmetic); I like Indiana University because of Lawrence Moss's work as well as the fact that you can have an additional concentration in cognitive science; I like U Pitt because I could work with the CNBC people; I like Stony Brook because of their strong algebraic geometry department, the list goes on for several more schools, and again this is just for math, I'm also interested in applying for statistics and at least a couple cog sci places.

I'm still wondering about the idea of contacting professors.

 

[deRham]

You should apply to as many schools as you would be happy attending. I would expect that to be 5-10.

That is extremely low a number. Perhaps you are advising with physics in mind, but in mathematics, it's unlikely that you will find only that many places that are doing what you're interested in. Also, it greatly messes with your chances of getting in somewhere. I would say more in the 10-15 range is safe in the sense that you develop a pretty good chance someone you really like will accept you.

Unfortunately, that is costly, so it might become necessary to apply to fewer, but I would certainly not recommend stinginess here, because it's your future. If you do apply to very few, I would recommend, keeping nearly a third of them schools that are very easy to get into that you can very well see yourself getting into, but still with people you can think of working with. Definitely don't make those five MIT, Harvard, Princeton, ...

I am considering contacting a couple of professors at each school who do work I find interesting, would this be recommended? Should I inquire as to the likelihood that they won't have too many students they will be advising should I be accepted to the program, or is that gauche?

I think it's fine if you ask. Every student's needs are different, and a non-snarky professor often will be happy, as long as you keep your mails to the point and respectful. Also, check if they even will be taking students - not all professors are going to be for some time.

 

[Vanadium 50]

Fair enough. Yes, in physics 5-10 is more reasonable. There are only ~150 places that give PhD's in physics.

 

[Bourbaki1123]

That is extremely low a number. Perhaps you are advising with physics in mind, but in mathematics, it's unlikely that you will find only that many places that are doing what you're interested in. Also, it greatly messes with your chances of getting in somewhere. I would say more in the 10-15 range is safe in the sense that you develop a pretty good chance someone you really like will accept you.

Unfortunately, that is costly, so it might become necessary to apply to fewer, but I would certainly not recommend stinginess here, because it's your future. If you do apply to very few, I would recommend, keeping nearly a third of them schools that are very easy to get into that you can very well see yourself getting into, but still with people you can think of working with. Definitely don't make those five MIT, Harvard, Princeton, ...



I think it's fine if you ask. Every student's needs are different, and a non-snarky professor often will be happy, as long as you keep your mails to the point and respectful. Also, check if they even will be taking students - not all professors are going to be for some time.

Thanks, this is the sort of information I was looking for.

 

[deRham]

You're welcome. And you seem to have a very good idea of things you might want to do. I have deceived myself many times about this sort of thing. I think once people really dig deep into what could please them in research, they will discover interest in many schools.

I hardly have broad interests.

 

[Stengah]

Just came across this on the US News website about graduate schools. The article was discussing engineering grad school, but close enough. This is from the "Getting In" section.

Introduce Yourself. Since admissions are often decided among faculty rather than admissions officers, try to make contact with a potential adviser in your desired department. Start off with a "snail mail" introduction that includes a brief bio, a photo, and an explanation of why you'd like to work with that professor, recommends Angela Lindner, associate dean at the University of Florida. Then, follow up via E-mail.

http://www.usnews.com/education/articles/2010/04/15/engineering-schools-tips-and-stats

Anyone have an opinion on this? I didn't think about snail mail, shows a little more effort than email.

 

[Bourbaki1123]

Just came across this on the US News website about graduate schools. The article was discussing engineering grad school, but close enough. This is from the "Getting In" section.

http://www.usnews.com/education/articles/2010/04/15/engineering-schools-tips-and-stats

Anyone have an opinion on this? I didn't think about snail mail, shows a little more effort than email.

I don't know if I would feel comfortable doing that. Maybe it's an age gap, but I would personally be more likely to be a bit uneasy about receiving that sort of package rather than an email expressing some kind of detailed interest were I in the professor's position, the detail of enclosing a photo is particularly strange to me.

 

[deRham]

I suggest avoiding all that and just e-mailing them. Professors are most used to that right now, at least from what they seem to tell me. It's perfectly reasonable to send a to the point e-mail.

Of course, some professors are magical at responding (despite their insane schedules), and others simply aren't able to handle all their e-mail.

 

[Stengah]

Yeah I thought the photo part was a bit creepy, but I do know some professors that are pretty anti-computer and would probably rather get snail mail. Everyone is just different though.

 

[Vanadium 50]

However, if it looks like you are spamming a zillion places, this will certainly be viewed negatively.

 

[Bourbaki1123]

Yeah I thought the photo part was a bit creepy, but I do know some professors that are pretty anti-computer and would probably rather get snail mail. Everyone is just different though.

I don't think I'd really want to work with a prof who is anti-computer.

 

[Bourbaki1123]

However, if it looks like you are spamming a zillion places, this will certainly be viewed negatively.

Of course, that's why you have to talk about their interests and what it is that you like about their work. Other than that, I doubt it will be viewed as spam unless I start emailing multiple people in the department (which I don't plan on doing).