This National Science Foundation (NSF) Project Grant award under the Computer and Information Science and Engineering program (CFDA 47.070) is for $291,261 over 3 years from January 1, 2024 to December 31, 2026. The grant is funding collaborative research by the Illinois Institute of Technology (IIT) to develop efficient graph neural network (GNN) algorithms and computation systems for both static and dynamic graphs.
The key products and services to be delivered include:
Design of novel, efficient GNN algorithms that can retain the entire graph and offer convergence guarantees
Development of efficient GPU-based computation techniques with optimized workload scheduling and reduced synchronization overhead
Novel strategies to address algorithm and computation challenges for GNNs on dynamic graphs
Creation of a high-performance software library as a foundational tool for the research community
Integration of the research findings into graduate and undergraduate curriculum
Outreach and educational activities to promote participation of underrepresented groups