Date: Monday, October 20th, 2025
10:00 am – 11:00 am Pacific Time
1:00 pm – 2:00 pm Eastern Time
Location: Weekly Seminar, Zoom
Title: : Imbalance Tilts the Scales: Assuring Fair Performance
Abstract:
Critical industries such as medicine and finance are built on inherently imbalanced data and prediction tasks. In this talk, we explore the fundamental limits of the two most common forms of correction: downsampling and upweighting. We present theoretical insights into the regimes where they work as expected, where they fail, and how they can be adapted to meet the needs of modern, highly overparameterized systems. Additionally, we detail algorithms to ensure robustness for both discriminative and generative systems trained on highly imbalanced data.
Bio:
Lalitha Sankar is a Professor in the School of Electrical, Computer and Energy Engineering at Arizona State University. She received a bachelor’s degree from the Indian Institute of Technology, Bombay, a master’s degree from the University of Maryland, and a doctorate from Rutgers University in 2007. Following her doctorate, Sankar was a recipient of a three-year Science and Technology Teaching Postdoctoral Fellowship from the Council on Science and Technology at Princeton University, following which she was an associate research scholar at Princeton. Prior to her doctoral studies, she was a senior member of technical staff at AT&T Shannon Laboratories.
Sankar’s research interests are at the intersection of information and data sciences including a background in signal processing, learning theory, and control theory with applications to the design of machine learning algorithms with algorithmic fairness, privacy, and robustness guarantees. Her research also applies such methods to complex networks including the electric power grid and healthcare systems.