Instructions

Prospective Undergraduate Students:

• If you are an undergraduate student interested in working with us, you must meet the following requirements:

• Able to work at least 5-10 hours a week

• Self-motivation and an interest in researching reasoning

• Willingness to do the kind of work few others in the world are doing

• An ability to tackle new and difficult problems in creative ways

• Ideally: Some background (either through coursework or your own independent studies) in one of the topics we research.

• Strong preference for students who have already taken: data structures and algorithms, an AI class, and a machine learning or NLP course (if you haven't yet been able to take one at USF, a good idea is to work through the Stanford NLP course online).

• Students who have been working with me a few semesters and have proven themselves capable of research may be eligible for paid research positions, and the chance to be involved in research publications. I also offer for-credit options, through independent study or supervised research. If you are interested, please take a good look at our areas of research, and contact Dr. Ankur Mali with a CV or résumé that clearly shows your relevant experience, interests, and coursework.

• How to participate. If you're a good fit, then there are at least two ways you can participate:

• Volunteer. You can join the lab as a volunteer, in which case your obligations are minimal. But I'm also very selective about who I take on as a volunteer. If it appears that you're going to be a student who just sits silently in lab meetings and doesn't do much else, then I'm sorry, but you're not a good fit.

• Course Credit. You can sign up for an independent study or supervised research credit; these are 3 credit courses offered through the CSE department (note that there are course prerequisites for these).

• Things to NOT do: I often see students making the following mistakes in contacting me. Try to avoid these:

• Contact me without reading this site thoroughly. I'm looking for people who can be independent researchers, and if you can't show that you're capable enough to find this website and read it (or worse, you ask me what the URL is), I won't have confidence you'll be a good researcher.

• Expect me to do your work for you. I want to see that you have taken the time to read my papers, really understand what my lab is trying to do, and figure out how you can contribute. I do not have the luxury of taking on students whose hands I have to hold with every little task. I'll guide you, but I won't drag you along.

• Use general platitudes. If you find yourself writing an email that says "I see you do a lot of research in AI, and I think it's interesting" and not much more, it's a huge red flag that you didn't actually take the time to understand what my lab is working on. That doesn't bode well for my confidence in your abilities.