UCLA Samueli
Sep 1, 2022
UCLA Samueli Newsroom
Baharan Mirzasoleiman, an assistant professor of computer science at the UCLA Samueli School of Engineering has received a National Science Foundation CAREER award — the agency’s highest honor for faculty members in the early stages of their teaching and research careers. The award includes a five-year, $529,000 grant to support Mirzasoleiman’s research to improve efficiency and robustness of machine learning from massive data sets, with sustainability as a key goal.
Machine learning — a branch of artificial intelligence using computers to sift through massive and complex data sets to learn patterns and offer insights — has become a ubiquitous technique for a wide range of applications in health care, science, technology and business. But while machine learning algorithms have continuously been improved to enhance performance, the energy used and its resulting carbon footprint has also increased.
With the award, Mirzasoleiman will explore new ways to build powerful machine learning algorithms that, rather than training on the entire data set, will instead focus on relevant core subsets capable of offering the range of complexity needed for accurate, robust, and efficient training, thereby cutting down on the time and computing power required. The resulting algorithms will be broadly applicable for learning from massive data sets across a wide range of applications, such as medical diagnosis and environment sensing.
Mirzasoleiman joined UCLA Samueli in 2020. Her research in machine learning from massive data sets has applications in image collections, recommender systems, web and social services, as well as video and other large data streams. Prior to UCLA, she was a postdoctoral scholar at Stanford University. Mirzasoleiman received her Ph.D. from the Swiss Federal Institute of Technology in Zurich, Switzerland, where she received an award for an outstanding doctoral thesis. She has also been selected by MIT as a Rising Star in Electrical Engineering and Computer Science.