Research opportunities in Theory?

[ahsanxr]

So is anyone else having a hard time planning their summer in terms of research just because they're only interested in doing theory research over the summer (physics). I definitely am. So far the only option I've come across is a Professor who seems rather unwilling but has agreed that if I do an independent study with him, we could work something out. The problem is I probably won't have much time to devote to this independent study and even if I did, from the looks of it, it doesn't seem like it'll work out very well. As far as REUs go, I've seen a couple of ones with interesting theory stuff, but those are few and far between, so I probably shouldn't count on getting in. Who else is in this situation and how are you guys coping? Do you guys know of any programs?

It's my 3rd year and I've taken a couple of graduate courses in physics, but I haven't done any research yet, so I definitely want some on my application before its time to apply for grad school. Alternatively, how bad would it seem for a theory applicant to not have done any research?

 

[Jorriss]

Why do you want to do theory? Have you considered computational or experimental work?

 

[ahsanxr]

Why do you want to do theory? Have you considered computational or experimental work?

I'm not interested in experimental work. I guess if I can find nothing else, computational would be fine too, but I haven't programmed in a while.

But if one were to ask what my interests were, the answer would be mathematical physics (i.e rigorous formulations of existing theories) or physics that uses a lot of abstract math (GR, string theory, unification etc). So that's what I'm going to be shooting for in grad school. Obviously, I don't have the background to work in something like this yet, but I still need something to put on my apps so at this point I'd be happy doing something in theory in general.

 

[NewtonsHead]

Shoot for the stars. Good luck man.

 

[rhombusjr]

A lot (most?) of theory work is computational. It's hard to tell if your theory matches experiment without running some simulations.

For analytical math type theory work, do you know any differential geometry? It's essentially a pre-req for what you want to do, plus it has applications to things you already know, like E&M, classical mechanics, and thermo. Teaching yourself math is probably the only thing you can do if you don't want to do computation or experiment.

I would be wary of being so dismissive about computation and experiment. Even though they're not exactly what you want to do in graduate school, both of them will give you good things to put on your apps. You'll note that most schools make theorists gain experimental experience anyway, so wanting to be a theorist but having an experimental background still looks pretty good.

 

[ahsanxr]

A lot (most?) of theory work is computational. It's hard to tell if your theory matches experiment without running some simulations.

For analytical math type theory work, do you know any differential geometry? It's essentially a pre-req for what you want to do, plus it has applications to things you already know, like E&M, classical mechanics, and thermo. Teaching yourself math is probably the only thing you can do if you don't want to do computation or experiment.

I would be wary of being so dismissive about computation and experiment. Even though they're not exactly what you want to do in graduate school, both of them will give you good things to put on your apps. You'll note that most schools make theorists gain experimental experience anyway, so wanting to be a theorist but having an experimental background still looks pretty good.

I know a little manifold theory and differential forms (and the symplectic formulation of CM). Riemannian geometry, which is probably the most applicable part though, I'll be learning this semester through my GR class. And yes, you're correct. It seems like most analytic work involves this type of math (topology and geometry). Even the condensed matter theorist I spoke with and mentioned in my OP was mentioning K-theory.

Sure, experiment/computation is better than nothing but before doing that, why not try getting something in your area of interest? Secondly, it would probably affect my performance as well if I'm doing something I don't care much for.

PS, speaking of applying forms to thermo, does anyone know of a book that introduces thermodynamics, but uses the formalism of differential forms?

 

[Lavabug]

Keep an open mind, science is still science whether you're doing it on the blackboard or soldering to a PCB board. Feynman could repair radios and assembled the old IBM computers they used in Los Alamos without an instruction manual. Fermi and Lorenz, both also known popularly as theoreticians did tons of experimental work.

I've always thought computational physics = theory, just the kind of theory you can't realistically do by hand (N-body simulations, numerical relativity, electronic structure, molecular dynamics etc). So you better get to like it, you may very well end up needing much use of it in any theory phd.

 

[cgk]

I second Lavabug. Computation is part of theory. One of the pillars of physics is knowing how to approximate things --- and that is also the core of numeric computation. Also note that there are areas in theory where people won't even bother looking through your article unless you prove, numerically, that whatever you tought out works at least in some cases (and even areas where they won't look if what you constructed does not works better than /everything/ else in at least some cases).

You should not ditch experiments and computation. They are important not only in themselves, but also for doing good theory.

 

[ahsanxr]

Sure, experiments and computations are awesome. But what's wrong with having a preference? And then acting in accordance to those preferences?

Firstly, from what I can tell, computers aren't employed that much by people who work in my areas of my interest. Secondly, I'm not against computers or anything. I know C and could probably pick the stuff up if I ever need it.

Lastly, why do people completely ignore the original question and give their advice instead? Your opinions are welcome but please also try to address the original question.

 

[micromass]

Have you tried approaching mathematics professors?? Some mathematics professors are interested in physics and might not mind to do research with a physicist. Furthermore, doing research in pure mathematics (without a link to physics) might also not be bad: at least you'll learn the relevant mathematics.

This might be something to look into. However, I don't know how physics grad schools look at pure math research.

 

[Lavabug]

Sure, experiments and computations are awesome. But what's wrong with having a preference? And then acting in accordance to those preferences?

Firstly, from what I can tell, computers aren't employed that much by people who work in my areas of my interest. Secondly, I'm not against computers or anything. I know C and could probably pick the stuff up if I ever need it.

Lastly, why do people completely ignore the original question and give their advice instead? Your opinions are welcome but please also try to address the original question.

Because the fact of the matter is: there are fewer opportunities to be found doing clean, pen and paper work then there for doing more practical hands-on projects. Some research is better than no research in the eyes of any graduate committee, regardless of what it was.

 

[rhombusjr]

You mentioned GR as one of your interests. A lot of the theoretical work in GR is labeled as "Numerical Relativity." It is highly dependent on computation. In string theory &c., you can't calculate anything, so nobody does. Also, string theorists seem very passionate about having a theory which doesn't match up to anything in real life, so why would they care about seeing if their theories match experiment?

What do you expect to find in your purely analytical research? In computation, you can see how a theory predicts nature to behave. In experiment, you can see how nature actually behaves. What do you expect to get out of analytical math?

Here is an example of some mathematical physics work you can do (if you haven't done so already): Compare solutions to Maxwell's equations in tensor/vector calculus vs. differential geometry. What's different? Can you say things about physics from the DG formulation that you can't infer from the vector formulation? Vice-versa?

Like others have said, try looking in the math department. A lot of string theorists end up there.

Most schools will expect applicants not to have done any theory research. Undergraduates typically don't have enough knowledge to meaningfully contribute to theory work except as a simulation programmer. The important thing is that you get some sort of experience. Most schools aren't too concerned about the specifics of your research, as long as you've done some.

To specifically answer your original question: It will look pretty bad. The important thing is to get research experience regardless of what type. It's fine to have a preference, but don't close doors because you're too focused on something else.

 

[SophusLies]

I'm not interested in experimental work. I guess if I can find nothing else, computational would be fine too, but I haven't programmed in a while.

But if one were to ask what my interests were, the answer would be mathematical physics (i.e rigorous formulations of existing theories) or physics that uses a lot of abstract math (GR, string theory, unification etc). So that's what I'm going to be shooting for in grad school. Obviously, I don't have the background to work in something like this yet, but I still need something to put on my apps so at this point I'd be happy doing something in theory in general.

I'll just give some experiences so you know what you're getting into. I worked for about 4 years after I got my undergrad then I went to grad school so I got to see my other classmates that went straight to grad school graduate and get jobs or not. I've seen several people in fields like the ones you're interested in not get jobs and most didn't even get post-docs. These were the students that were good and I mean real good and also went to very good schools.

Before I officially picked my research adviser I asked professors of many fields what their alumni are doing. The hardcore theory alumni don't get jobs in those fields and most don't even get jobs related to physics. I'm only telling you this because I wanted to do mathematical physics 100% when I was thinking about grad school but that quickly changed when I found out it might cut me out of the physics field completely. If you like theory then make sure it has some sort of computational element so if things don't work out, which they probably won't, you can get into something related to physics or math if that's where your heart is.