1

I am trying to calibrate my expectations for PhD applications in computer science / AI (with an interest in machine learning and computer vision) and would like to understand how to identify realistic programs rather than just aiming blindly at the most competitive universities.

My profile in short:

  • Master’s degree in artificial intelligence (Europe), completed recently with top grades and honours.

  • Master’s thesis in modern computer vision / 3D representation learning (e.g. neural radiance fields).

  • Several substantial AI / vision course or side projects (implementing and extending existing methods).

No research publications so far.

Bachelor’s degree grades are average compared to my master’s performance.

Geographically, I am mainly considering Europe and Canada, and I need funded positions (stipend or salaried PhD; self-funding is not an option). After some rejections from highly competitive programs, I am trying to understand how to choose my targets more realistically.

My questions is:

  • What objective criteria can applicants use to judge whether a PhD program is a realistic target (beyond general prestige)? How can you judge what program is better than another program?

I am not looking for recommendations of specific universities or rankings, but for general criteria and approaches that applicants in a similar situation can use to evaluate programs and set realistic expectations.

Improve this question
5
  • It sounds like you potentially have a lot of research experience and not all of it is so recent as to not yet be ripe, why hasn't it led to any publications or conference presentations? Commented Nov 19 at 22:04
  • @BryanKrause In my master’s program I completed several research-oriented projects, but I did not receive close supervision regarding how to turn them into publishable work. My thesis was primarily overseen by a PhD student, and I had limited contact with the professor, so I did not receive guidance on publication venues, drafting a paper, or preparing a submission or getting mentored to do research. As a result, although I worked on research-type topics, none of the work progressed to a formal publication and also recive the letter of recommendation of my thesis supervisor. Commented Nov 19 at 22:15
  • What makes a masters record strong is the potential for publications. Usually that is what separates undergraduate vs masters level applicants, the exposure to research and the ability to accumulate publications. At this point, in my opinion, the masters without output is neutral. Commented Nov 21 at 20:43
  • It's not even productive to say to do a PhD you need a publication. In bachelor's and master's no one teaches you how to do research. You could explore research only in bachelor's and master's thesis and it depends on the professor. Imagine you choose a professor who doesn't mentor you and let you work with a PhD student. It's often for most professors to not even mentor the student to do research. My professor didn't even show up for my graduation and it's a well know researcher, or at least he publishes on top conferences. Commented Nov 23 at 19:16
  • @Nobodyet The problem for you is that the PhD candidates in AI that are being accepted now, do have publications, and they did receive the mentoring to be able to write or collaborate in publications. So it is a different standard, and there are Master degrees where they teach you how to do research. Commented Nov 25 at 10:28

0

Reset to default

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.