The National Science Foundation (NSF) Division of Mathematical Sciences awarded Yale University a $159,979 Project Grant under the Mathematical and Physical Sciences program (CFDA 47.049) to advance robust machine learning methods for classifying data and quantifying prediction uncertainty in complex, non-IID settings. The project aims to develop innovative classification strategies that exhibit improved worst-group performance across latent sub-populations and enhanced fairness. These strategies will integrate with state-of-the-art techniques like neural networks and gradient boosting. The project will also investigate adaptive classification methods that leverage both labeled training data and unlabeled test samples, with the goal of facilitating application in safety-critical and fairness-critical domains like medical diagnosis. The grant term runs from September 1, 2023 to August 31, 2026.
Generated 4/30/24, 8:25 AM