I am not sure if it is the right place to ask such a question, but I don't know any other place that would provide a serious answer or advice for me.

I am an undergraduate student (and a Sophomore starting in September this year) in Logistics field. Before entering the University, I wanted to study in the field Computer Science and Engineering (CS&E) which is offered at my University. But due to some reasons, (one of which was a fear of failing at Physics to enter CS&E) I had to apply for Logistics field and transfering students from a field to a field is not offered at the University.

I really want to get a PhD in Artificial Intelligence (I am very serious about it), but considering that I am studying in a field completely unrelated to Artificial Intelligence, I am being depressed and worried that I wouldn't be able to get a PhD in AI.

But I still want to ask: will I be able to get a PhD in AI field? I haven't yet stopped my habit of learning higher Mathematics, Electronics, and I am planning to read plenty of books on Computer Science.


If this is the US and you want to study in the US, then it might be possible, but the only way to be sure is to put together a good application and send it to a few places. They will help you understand any specific missing pieces needed before you can start.

But the nature of both undergraduate and graduate education in the US can make this possible. The US undergraduate program is very general, with students studying many things besides their "major" subject.

Graduate admissions programs recognize this and also seldom require a masters before starting a doctorate. They therefore compensate by requiring (more or less) certain advanced courses, culminating in required qualifying examinations before you start serious work on research.

Having little CS background, though, may be a block at some places.

But, in general, your application needs to show that you are a good candidate for success in the program. That has many elements, not just the undergraduate major.

In other countries, say UK, this would not be the case and it would probably be much more difficult. There, a much deeper understanding of the subject is required at entry.

|improve this answer|||||
  • 3
    If the OP plans to do the AI PhD in a computer science department, they may be expected to have the programming skills of a good CS graduate. That is not just a matter of reading computer science books, but of practicing programming. – Patricia Shanahan Jun 10 '19 at 12:58
  • 1
    @PatriciaShanahan, true enough. And (a) programming is a skill held by many and (b) programming is only a tiny part of a CS degree. But if you can program you can build interesting things. You don't need a CS degree to be a skilled programmer. – Buffy Jun 10 '19 at 13:04
  • 2
    Programming is a tiny part of some CS classes, e.g., ML, which is mostly linear algebra. But it's a big part of most UG CS classes -- or at least, it is here at Michigan. But also, I think Patricia is making the point that it's not enough to know the material. A successful applicant also has to demonstrate they know it and can apply it. That's usually achieved by getting lots of A's in relevant courses and getting some strong LORs from instructors in those courses. Claiming you've read a lot of books on your own isn't really a substitute. – Nicole Hamilton Jun 10 '19 at 14:38
  • 1
    Agreed that the student needs to take classes in CS, even if they're not counted towards their major. – Parever Jun 10 '19 at 18:03
  • 1
    Actually, @NicoleHamilton, I wasn't speaking about classes, but about the program as a whole. Students don't take the compiler course, for example, to demonstrate programming a compiler. That would be too shallow a view. And I don't understand the difference between "know the material" and "demonstrate". They are, to me, the same. If you can't do the latter you haven't achieved the former. Yes, students in CS do a lot of programming but it isn't so that they can just be programmers. – Buffy Jun 10 '19 at 18:08

Impossible to answer without knowing more about your school, the curriculum in your present major and you.

Whether you could be accepted into PhD program in CS depends on what courses you take as an undergraduate, any research experience you gain along the way, your grades and the strength of your references. Even if you're not a CS major, can you take CS courses at your school or, perhaps, do a CS minor? If you're able to create a record that indicates you've developed basic competence in CS, that you did well in that coursework and could be a successful researcher, then, yes, it's possible.

If this isn't an option at your school in your present major and you can't change your major, then maybe you should consider transferring to another school where you can follow your heart.

But I'm concerned by your fear of failing physics. Physics is a lot of math and if you're afraid of math, this could be a big problem if you're interested in AI and, especially, machine learning, which is all math.

|improve this answer|||||

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.