Applicants ought to be highly self-motivated, with strong skills and interests in mathematical problem solving and/or experimental work, ideally both. Interviews will likely involve questions on probability, statistics, algorithms, programming, and machine learning.
We are particularly interested in people who can formalise and reason rigorously about the often murkily defined challenges facing modern AI, and work toward solving them in ways that are both principled and scalable.
Our research interests span several areas within responsible AI and learning theory, including robustness, online learning, privacy, machine unlearning, AI safety, and learning-theoretic perspectives on responsible ML.
We are looking for self-motivated PhD students to join the group in 2027, with strong problem-solving skills.
If you are interested and believe you would be a good match, please fill in the contact form. If your application is relevant, Amartya will reach out with more information.