This $249,999 federal Project Grant award from the National Science Foundation's (NSF) Mathematical and Physical Sciences (CFDA 47.049) program to Rensselaer Polytechnic Institute (RPI) supports research to develop novel mathematical models and efficient algorithms for deep learning on large-scale graph-structured data. The project aims to advance techniques for graph convolutional networks to enable accurate and efficient solutions for challenging graph tasks such as graph matching and graph clustering. Key research activities include designing variance-reduced neighbor sampling approaches, new constrained optimization models, and accelerated low-order methods to solve the new models. This research is expected to produce innovative mathematical models and algorithms that can significantly improve the performance of deep learning on large-scale graph data across numerous applications. The project will also engage graduate and undergraduate students, particularly underrepresented STEM students, in the research activities and integrate the findings into curricula to impact both graduate and undergraduate education.
Generated 1/28/25, 7:52 AM