Muthukumar Lands NSF Career Award for Foundational Machine Learning Research
Vidya Muthukumar has been named as a recipient of the National Science Foundation (NSF) Faculty Early Career Development Program (CAREER) award. Muthukumar is an assistant professor in the Georgia Tech School of Electrical and Computer Engineering (ECE) with a joint appointment in the H. Milton Stewart School of Industrial and Systems Engineering.
The NSF CAREER award is the most prestigious award in “support of early-career faculty who have the potential to serve as academic role models in research and education and to lead advances in the mission of their department or organization.” Approximately 500 awards are given annually to universities and research institutions throughout the country. The NSF especially encourages women, underrepresented minorities, and persons with disabilities to apply.
Muthukumar’s NSF CAREER project, "Overparameterization in modern machine learning: A panacea or a pitfall?,” aims to establish foundational principles — through a diversity of mathematical techniques spanning signal processing, information theory, and online decision-making — that explain the successful generalization of modern machine learning and its failure modes in order to develop efficient and principled solutions. In the absence of such a foundation, outstanding failure modes in deep neural networks remain unmitigated or unnecessarily costly to solve, and architecture selection is conducted in a wasteful trial-and-error manner that involves repeated train-and-test cycles. Muthukumar hopes that the outcomes of this project will eventually enable deep learning technology to reach its full potential in high-stakes and resource-limited applications.
Additionally, the project will create and disseminate educational resources at the high school and undergraduate levels on elementary signal processing, machine learning, and data science that underlie and complement the described research.
Before joining Georgia Tech in January 2021, Muthukumar earned her Ph.D. degree in EECS at UC Berkeley and spent a semester at the Simons Institute for the Theory of Computing as a research fellow for the program “Theory of Reinforcement Learning.” She is the recipient of an Amazon Research Award, Adobe Data Science Research Award, Simons-Berkeley Research Fellowship, IBM Science for Social Good Fellowship, and the UC Berkeley EECS Outstanding Course Development and Teaching Award. She currently serves on the senior program committee for the Annual Conference on Learning Theory and on the organizing committee for the Learning Theory Alliance mentorship organization.