My main areas of study are distribution testing (and, broadly speaking, property testing), learning theory, and, more generally, randomised algorithms and the theory of machine learning. One of my current focuses is on understanding the computational aspects of learning and statistical inference subject to various resource or information constraints. Another, not quite disjoint from the first, lies in reliable and rigorous approaches to data privacy, specifically differential privacy.
I am currently looking for Ph.D. students! If you are an undergrad/masters student with a strong background in algorithms and/or discrete mathematics interested (broadly) in the theoretical aspects of learning, and are keen on spending 3-4 years in one of the world's best places to live, please contact me, including your CV! The position is fully funded, and involves proving (or trying to) theorems about the fundamental limits and possibilities of learning and testing algorithms, from a computational and statistical perspective. Please check my publications for some recent work along these lines.