I have multiple openings for graduate students at all levels.


Anyone entering the academia should internalize two virtues: Academic Integrity, and Intellectual Humility. In other words, you should be 100% honest about what you know and what you don’t; what is absolutely true and what is likely to be true; what you learned from others and what you discovered; what you understood, and what does not make any sense to you.

In addition, I am looking for students with following strengths:

  • Work ethics: Graduate school is a 5+ year journey with very sparse rewards. Even the students with strongest motivations often get lost. “Pushing forward no matter what” helps you get through the uncertainties.
  • English proficiency: If you can’t read, you cannot find a good research problem. If you can’t write, you cannot convince people of the benefits of your solution.
  • Mathematical foundations: If you are theory-inclined, I need not say any further. If you are algorithm-inclined, math should help you modeling the goal you want to achieve, and design straightforward solutions.
  • Programming skills: As a bare minimum, you need these to validate your idea.1 Furthermore, a well-written code can save you a lot of time.

Lacking one or two of these could be okay, if you are ready to work hard enough (but you still need a B.S. degree).


I wish this research group to be a place where we learn from each others' unique background, strengths and interests. Here’s the reason: The life in academia is a little bit of everything. You need to code, prove, write, draw, make videos, dig through ancient documents, persuade people to fund your research, et cetera. If you already have some research or work experiences (regardless of the field), that’s fantastic. If you are a math/coding guru, that would be excellent too. If you know how to write or design, I really want to learn a lot from you.


Step 1. Please send me an email with a CV (or equivalent) and the official transcript attached.

Step 2. You should apply to the EE@POSTECH directly before the deadline; there is a department-level screening procedure.

  1. This is extremely useful even for theoreticians, as you can often test if your claim is true, even before actually trying to prove it. Also, if you can show that your theory is actually useful in explaining the practice (which rarely happens in ML), your work can bring a humongous impact. ↩︎

Jaeho Lee

ML researcher who also teaches. jaeho-lee (at)