1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Senior in high school looking for most optimal path to a good CS grad school.

  1. Mar 19, 2010 #1
    The best I could possibly get in with all my effort (whether this is be the #5 CS grad school in the nation, or #40, does not really matter.. as long as I take the most optimal path there and do it ethically).

    I was wondering what aspects the best/better CS grad schools look for in a college graduate. Is it research? rigor of classes? GPA?

    The thing is, I was accepted into the McCombs school of Business at UT Austin for the fall semester as a freshman but I realize that I did not go with what my heart told me to do (to have applied for CS instead). My interests are in AI and machine learning.

    Now I have two options I guess:
    1) Apply as an internal transfer student to the CS department as a sophomore (no guarantees) or
    2) Continue with the 4-year business degree (MIS not offered), but do more CS research in the time leftover from this "easy" degree - I've heard that business majors usually have a lot of free time.

    What do you guys think about these two options? Is #1 the wiser way to go, #2 riskier?
    The problem arises from the fact that I have no familiarity with the system and thus no idea what good grad schools value the most in their applicants. whether it be upper-level CS courses or real work with a professor/postdoc.
     
    Last edited: Mar 19, 2010
  2. jcsd
  3. Mar 19, 2010 #2
    If you're interested in the business side any, CS and business can set you up well for an information management job. There are plenty of jobs to go for with just the business major too. CS electives will only help when applying to business jobs.

    For a CS PhD I don't know if the business side will help. Without the CS major it could certainly be tougher. Is there any way to double major if you can't transfer? The good thing about college is for the most part it's pretty simple... just keep your grades up. Usually you'll have a senior research project you can do, and there may be opportunities as a research assistant. Extra math will probably be required, and don't miss any opportunities for internship or research during the summer. You can start applying to some internships and summer programs starting freshman year.
     
  4. Mar 20, 2010 #3
    Do a BS in CS if you're very good at higher math, starting with Calculus. If you have any difficulty with higher math, you might still be able to get the degree, but your grades for the BS might not be good enough to get into a MS program.
     
  5. Mar 20, 2010 #4
    Let me try to clarify here. I chose business when I applied to the university because I thought accounting was a relatively safe, stable career and that computer science was somehow less so. I find computer science more meaningful but that doesn't mean I find every aspect of it to be "fun". It is work, like any other work but I find computer science the most meaningful compared to any other career/job out there.

    Also, I found that I value this meaningfulness in a career more than the supposed stability of accounting. Thus, if I choose to continue with a business major, it will not be for my original intent of pursuing accounting but maybe an easy, blow-off one such as marketing, finance, or entrepreneurship. This would (hopefully) free up much of my time for research experiences in Computer Science. This also raises the question, how much does a comp sci major's classes (even upper-level ones) actually aid him in real comp sci research? I wanted to find out if the compsci classes are really just hoops to demonstrate that you can get a decent GPA, while having very little to do with the area of research I'm interested in (AI and machine learning).

    So the dilemma is whether to go with the business degree and make the research aspect stand out for CS grad school, or do whatever I can to get into the CS program, maybe not have as much time to do research due to the rigor of the academics but still do some research on the side when time permits. I was thus wondering how grad schools would like either option lol. I'm not familiar with the system very much so this may seem a bit naive :(
     
    Last edited: Mar 20, 2010
  6. Mar 20, 2010 #5
    There are plenty of very meaningful business careers too. That may be something you'll learn, but either way you should go with what you are more passionate about. You will be more successful and enjoy your work more that way. Don't underestimate your business classes though. A top finance grad from McCombs, for example, will have a lot of meaningful and potentially lucrative opportunities after graduation. The career track certainly isn't safe and stable though - it's very competitive :smile:.

    For CS, your courses will very much help you and be important. You will have to take lots of math and computer science to be accepted to grad school and also to be successful. Surprisingly, you'll actually learn something in college. It's not just about checking the boxes. Your courses will have a lot to do with what you are interested in. Electrical engineering courses will also cover a lot about machine learning and AI, so check that out too.
     
  7. Mar 20, 2010 #6
    Thought I'd also throw in an example of what someone with a background in CS and business can do after graduation. General business type roles won't have anything to do with AI or machine learning though. The jobs involving those will be research focused, and the business background probably won't help you any.

    http://www.ge.com/careers/students/imlp/index.html
    http://www.imlpblog.com/
     
    Last edited by a moderator: Apr 24, 2017
  8. Mar 20, 2010 #7
    I'd go for the internal transfer. If you have no interest in Business, you will probably quickly become bored with your courses and either your GPA will suffer or you will ( having to force yourself to get through classes you don't give a crap about is pretty torturous).

    If you can, I would suggest trying to switch majors right away. I don't know about you but I couldn't stand being in any non-technical degree program (aside from philosophy, that is). I should issue a caveat; this is partly due to my political persuasions, and partly to do with my attitude towards corporate life (which I may well have to swallow in the future).

    The biggest problem will likely be that (this is really just from my personal experience) most of the people will probably have a significantly different mindset than you. Having little peer support (in the sense that other people share your aspirations and interests) may be very discouraging for someone with aspirations to go into the pure research side of things.

    The most pertinent question raised in this thread is that of your math ability (though I don't see how you could do an accounting major without at least some, so I would guess that you are good on that count). I couldn't tell you how many people I've seen switch majors and/or end up going for six years because they simply couldn't hack the math requirements. If you suck at math, you will have some problems with CS and you will almost definitely have some problems at the graduate level.

    The mathematics done in CS research tends to be of a very different feel than that done in say, Calculus. I would say that upper level mathematics feels more like reasoning through a complex program than it does like reasoning through a calculus problem (of course this is entirely subjective). Nevertheless, you have to be able to get through calculus to get your degree. I can tell you for a fact that you will refer back to (among other things) the infinite sum stuff you will do in second semester calculus when you are looking at algorithms in AI.

    Also, the professors you will want to do undergraduate research with will usually be the ones teaching the upper level courses you should be taking. They will probably be apprehensive about taking up a business major on a CS project. So in reality grad schools value both these things, as getting research opportunities is usually contingent upon having the experience/knowledge gained from upper division courses.

    So yes, change your major as soon as possible if you feel pretty solid about CS grad school.
     
  9. Mar 20, 2010 #8
    Since when? The closest I've seen in EE are the switching systems/digital logic courses, but that's a stretch.

    I'm jumping on the bandwagon with everyone else; switch into CS 'cause you won't have the theoretical comp sci background for AI/machine learning otherwise. Even a math major focusing on graph theory puts you closer to what professors will be looking for.

    A lot, actually. I don't use most of my EE background when playing with robots, but all those courses on algorithms, data structures, and operating systems actually come up when I'm trying to figure out the best way to solve some weird coding problem so that I can get a result in a sensible amount of time without causing the computer to freak out (I work with large datasets). You'll have to think about it too when you're actually coding up AI/ML type solutions, 'cause that's how you write good, robust, code that works well for your problem. CS courses just really play well together to help you become better at CS as a whole if you really pay attention.

    Carnegie Mellon has pretty comprehensive info on what they want in their applicants; I can't imagine other schools are very different.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Senior in high school looking for most optimal path to a good CS grad school.
Loading...