PhD Application Advice for Machine Learning and Artificial Intelligence

Here is a (curated) repository of high-quality resources that helped me understand the ML/AI PhD Application process and craft my application better.

This post orginally appeared on Twitter:

Are you applying for a PhD in Machine Learning, Artificial Intelligence, and beyond?

Here's a thread of high-quality resources that helped me understand the process + craft my application better. 👇

— Chaitanya K. Joshi (@chaitjo) October 3, 2021

Overview

If you don’t know where to start, I recommend Tim Dettmers' guide to ML PhD applications for an overview of the entire process: Link

Nelson Liu wrote a similar posts with a focus on Natural Language Processing PhD applications: Link

Other helpful articles with a broader scope include this one by a CMU Professor and this one from Stanford’s CS department.

Here is another helpful panel discussion on US PhD applications in ML/AI: Link

Personal perspective: ML/AI research is hyper competitive. I remember feeling extremely stressed out when comparing my profile to folks who were at the top labs and universities that I dreamt about. In those moments, it is important to remind ourselves that everyone’s path is different + why you love science & research + it will all work out for the best in the end! 🌞

Personal Statements and Research Proposals

MIT EECS has very useful and actionable advice on structuring PhD application essays, including annotated examples: Link

Personal perspective: For me, this part of the application process was extremely challenging but ultimately rewarding. Why? Well, it was probably the first time I wasn’t bouncing from one research project to another, and got to sit down and think about my long-term research vision, compress it into a coherent document, and share it with scientists.

P.S. Once you have prepared a draft, don’t forget to solicit feedback - you can ask your professors, connections, fellow applicants, etc.

P.P.S. I am happy to share my personal statements upon request.

Chosing Labs and Supervisors

This one-page document can serve as an excellent template when thinking about labs, supervisors, and what questions to ask when determining personal fit: Link

The CMU ML blog has comprehensive advice on going about choosing a supervisor and questions to ask them: Link

I also found this perspective from a professor at Stanford to be very insightful.

Personal perspective: Every PI may seem like a holistic and friendly person on social media or to established scientists. Please take the time to have a candid chat with current as well as former students to understand their personalities and the research group’s morale. You may think that the advisor has to pick you, but you are also picking the advisor!

Finally, if you need a cheerleader or a listening ear, if you are coming from an underrepresented background, if you are a south asian/south-east asian without strong connections and support, please do not hesitate to reach out to me or others around you. We are a research community. 🤗

May The Force be with you this application season! 💙💙💙