My research is in machine learning theory and algorithms. I focus mainly on active and interactive learning: a setting where the data source and the algorithm interact in an attempt to improve learning accuracy while decreasing information costs. Much of my work revolves around developing general-purpose algorithms that can be useful in many different applications. I am also interested in fairness in machine learning. Please see my personal webpage for a list of my publications.
Open positions
I am currently hiring Computer Science Ph.D. and M.Sc. students. Current graduate students can also apply for a research internship in my lab through Vector Institute. The next available time period for Vector internships in my lab is Summer 2025. Please consult the Vector Institute website for application deadlines.
I am looking for students who are interested in machine learning theory. If you are interested, you are welcome to email me. In your email, please let me know why you are interested in being a part of my team, and what position you are looking for. Please also attach your CV, transcripts and any other supporting material that you think might be relevant. Please indicate any previous research experience, even if it is unrelated to machine learning. Note that I will not follow links to external material, due to computer security reasons. Please attach any relevant files directly to the email.
Sivan Sabato is an Associate Professor at the Department of Computing and Software at McMaster University, a Canada CIFAR AI Chair and a faculty member at the Vector Institute. Prior to that, she was at the Department of Computer Science at Ben-Gurion University, following a post-doctoral fellowship at Microsoft Research New England. She received her PhD from the School of Computer Science and Engineering at the Hebrew University of Jerusalem. Sivan serves as an Action Editor for the Journal of Machine Learning Research and for Transactions of Machine Learning Research, and routinely serves as an Area Chair at Machine Learning conferences such as NeurIPS, ICML, COLT, ALT and AISTATS. She was also the Publication Chair for ALT 2021 and a Publication Co-Chair for ICML 2022 and 2023.
Related Courses
Machine learning algorithms can have any number of possible interactions with their environment. This course presents the foundations of machine learning analysis under a range of interaction types, including the classical PAC (Probably Approximately Correct)-learning framework , online learning, and active learning. In this course, students will learn foundational tools and core algorithmic approaches for each of these models. In addition, the course will include a study of recent research on interactive learning models.
