Stefanie Jegelka

Associate Professor, Electrical Engineering & Computer Science
MIT School of Engineering
Website

Stefanie Jegelka is an Associate Professor in the Department of Electrical Engineering & Computer Science (EECS), and a member of the Computer Science & Artificial Intelligence Laboratory (CSAIL), the Institute for Data, Systems, and Society (IDSS), the Center for Statistics and Machine Learning at MIT and the Operations Research Center (ORC). She has been awarded a Sloan Research Fellowship, a National Science Foundation (NSF) CAREER Award, a DARPA Young Faculty Award and faculty research awards by Google, Two Sigma and Adobe.

Professor Jegelka’s research is in machine learning, and spans modeling, optimization algorithms, theory and some applications. In particular, her group strives to answer three questions that help make machine learning more widely applicable. First, how can we best learn to mathematically represent combinatorial data such as graphs or subsets? This question arises in learning with networks, molecules, simulations, recommendations or when learning algorithms. Second, how can we obtain accurate machine learning methods when resources, e.g., high-quality labels (generated by humans or expensive experiments) are scarce? Third, how reliable are modern machine learning methods, and how can we measure and improve reliability?

The wide impact of machine learning comes with a responsibility. Having collaborated with experts from a wide range of backgrounds in research, teaching and leadership, Professor Jegelka is dedicated to promoting a responsible use of machine learning technology. This involves its impact on many axes, e.g., the deployment of machine learning to truly benefit society, science and the environment, awareness of risks, and the minimization of its resource footprint. Hence, she is excited to join a diverse set of experts in the MCSC.