Mahdi Haghifam
About MeI am currently a Distinguished Postdoctoral Researcher at Khoury College of Computer Sciences at Northeastern University, hosted by Jonathan Ullman and Hongyang Zhang. In August 2023, I completed my PhD at University of Toronto and Vector Institute where I was fortunate to be advised by Daniel M. Roy. My dissertation topic was Information-Theoretic Measures of Generalization in Machine Learning. I also received my B.Sc. and M.Sc. degrees in Electrical Engineering from Sharif University of Technology. Internships and Research VisitsDuring Summer and Fall 2022, I was a research intern at Google Brain (Differential Privacy Team) where I was extremely lucky to be mentored by Thomas Steinke and Abhradeep Guha Thakurta. I was also a research intern at Element AI (ServiceNow Research Lab) in Winter 2019 and Fall 2020 where I had the privilege of working with Gintare Karolina Dziugaite in the Trustworthy AI Research Program. In early 2020, I had the opportunity to visit the Institute of Advanced Studies at Princeton as a visiting student for the special year program on optimization, statistics, and theoretical machine learning. Research InterestsI aspire to do fundamental theoretical research that impacts practical problems. My main areas of focus are Differential Privacy in Machine Learning and Foundations of Machine Learning. Recent topics include 1) developing practical algorithms for privacy-preserving convex and non-covnex optimization, 2) role of memorization in learning, and 3) understanding surprising generalization properties of LLMs particularly for algorithmic tasks. For a complete list of my publications and preprints, please visit the Publications page. Reach Out!Feel free to reach out if you'd like to discuss research ideas aligned with my interests. Also, I'm happy to offer guidance and support to those applying to graduate programs, especially individuals who might not typically have access to such assistance |