This question is in spirit a very general one. When a person (say in CS) is choosing among multiple options for gradschool, which one would experience and prudence point towards? Along with research interests and all that, things to factor in would be the extra strain involved, to be honest.

Case in point:

  1. Choice A: PhD advisor in the field of computer vision and machine learning. My undergrad experience was mostly in image processing, with one instance of training a rather vanilla backprop neural network. So this would be pretty much be building on the base of undergrad research experience. Yeah, the school is a tough one, and my math background is rather sparse. lots of catching up to do on the maths front.
  2. Choice B: PhD advisor does work in graphics. mostly interactive graphics, lighting models and all that. These are topics i am vaguely familiar with, never having taken a graphics course during undergrad. Nor did any research project on that. Its a fresh topic for me, but I am not qualified enough to judge whether to pursue this or not.

2 Answers 2


If you start by thinking about what's safe as an undergrad/grad, I think you may experience (unpleasant) surprises down the road. Do what you like best! Grad school is not an easy path, and motivation is sometimes hard to maintain throughout, so the best way to help your chances of success is actually to do something that you enjoy (and hopefully will enjoy for the next few years).


Personally, I think grad school is as much about learning than it is about doing research. Don't be afraid to learn new things and expand into new fields - grad school is exactly the time do that. My main regret from my own graduate studies is that I focused too much on doing research and not enough on learning. Of course it requires hard work, but if you don't want to work hard I would say grad school is probably not the right choice for you.

As Fx writes in his answer, you should pursue whatever interests you the most. Of course it is also important to have a good advisor - so if you have no knowledge of the field you might like to ask around.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .