The University of California, Santa Barbara (UCSB) was awarded a $439,984 Project Grant by the National Science Foundation's (NSF) Computer and Information Science and Engineering (CISE) Program (CFDA 47.070) to conduct research on mitigating bias in machine learning algorithms. The goal of the 3-year project is to develop theoretical frameworks and principled algorithms to enhance fairness in both static and dynamic machine learning-based decision-making systems. The research will investigate the impact of data biases on the geometry of deep neural network models and design provably fair algorithms. Additionally, the project will develop models to guarantee fairness in dynamic settings where group imbalances and feature distributions evolve over time. The award supports undergraduate student engagement in the research and includes outreach activities for pre-college students.
Generated 1/28/25, 7:05 AM