Project Grant 2546201
- The National Science Foundation Division of Computing and Communication Foundations awarded the University of Delaware $579,891 on October 1, 2026, under the Computer and Information Science and Engineering program (CFDA 47.070) to develop mathematical and computational tools for modeling directed group interactions in complex systems. The project will create a unified framework for signal processing, structure learning, and neural network modeling on directed hypergraphs—network models whose...
- This three-year, $345,853 project grant from the National Science Foundation's Division of Computing and Communication Foundations will support the development of a new hypergraph signal processing framework via tensor representations. Funded under the Computer and Information Science and Engineering program, the collaborative research aims to advance tools for exploiting multi-way interactions in complex data using hypergraphs rather than solely pairwise relationships as in simple graphs....
- Federal Grant Award Summary The University of Delaware received a $287,125 project grant from the National Science Foundation's Division of Computing and Communication Foundations (CFDA 47.070) effective October 1, 2025 through September 30, 2028. This award supports theoretical research in Graph Signal Processing (GSP), a computational discipline that analyzes data signals residing on irregular graphs and networks rather than regularly arranged homogeneous spaces. The investigator will...
- The National Science Foundation Division of Computing and Communication Foundations awarded The University of Kentucky Research Foundation $420,000 on October 1, 2026, under the Computer and Information Science and Engineering program (CFDA 47.070) to develop mathematical and computational tools for modeling directed group interactions in complex systems. The project creates a unified framework for signal processing, structure learning, and neural network modeling on directed hypergraphs,...
- The National Science Foundation Division of Mathematical Sciences awarded a $240,002 Project Grant to the University of Delaware from September 1, 2021 to August 31, 2024 under the Mathematical and Physical Sciences (47.049) federal grant program. The grant supports research into Neural Network Learning for Graph Data. The University of Delaware will develop methodologies, theoretical frameworks, and perform simulation and experimentation to advance the mathematics and science of graph neural...
- The National Science Foundation Division of Computing and Communication Foundations awarded the University of California, Santa Barbara $200,000 on September 1, 2026, under the Computer and Information Science and Engineering program (CFDA 47.070) to develop foundations for discovering higher-order scientific laws with deep learning. The award funds research on extending graph neural networks to model higher-order interactions in physical systems—interactions among components that extend...
- The National Science Foundation Division of Computing and Communication Foundations awarded Yale University $596,523 on June 1, 2026, under the Computer and Information Science and Engineering program (CFDA 47.070) to develop non-Euclidean representation learning methods for graph-based foundation models that improve artificial intelligence's ability to model complex network-like relationships in real-world data such as biological networks and drug discovery applications. The project addresses...
- The National Science Foundation Division of Computing and Communication Foundations awarded $499,928 to the College of William and Mary on July 15, 2026, under the Computer and Information Science and Engineering program (CFDA 47.070) for a CAREER project on graph dynamics in learning systems. The award funds development of graph learning algorithms that model and forecast behavior in dynamic connected systems—including coastal sensor networks, manufacturing workflows, and power grids—while...
- The National Science Foundation Division of Computing and Communication Foundations awarded North Carolina State University $300,000 on September 1, 2026, under the Computer and Information Science and Engineering program (CFDA 47.070) to develop methods for generalizable editing of foundation models, including large language models. The project investigates knowledge augmentation frameworks that enable foundation models to update stored facts more reliably and apply edits across related...
- The National Science Foundation Division of Computing and Communication Foundations awarded Old Dominion University Research Foundation $152,143 on October 1, 2025, to establish and operate a Research Experiences for Undergraduates (REU) site in graph learning and network analysis hosted by the Department of Computer Science at UNC Greensboro. The site will provide research training to ten undergraduate students each summer in the foundations of graph machine learning and network analysis and...
The National Science Foundation Division of Computing and Communication Foundations awarded the University of Delaware $597,508 on September 1, 2026, under the Computer and Information Science and Engineering program (CFDA 47.070) to develop theoretical foundations of hypergraph learning and characterize core features of hypergraph neural networks. The research will establish theoretical underpinnings for hypergraph learning methods that generalize standard graph learning by allowing edges to connect multiple data points simultaneously. The project will characterize two core features: the sample complexity required to achieve desired prediction accuracy and the model architectures needed for specific signal sources. Work will include studying expressiveness of hypergraph neural networks with tensor-based message passing mechanisms, developing discriminative models using hypergraph pooling and autoencoders, and exploring neural architectures' ability to discriminate between different graph structures to perform strong isomorphism tests. These methods address complex multi-way interactions in applications including social networks, gene-disease analysis, and brain functional analysis. The award supports curriculum design and outreach activities to broaden participation across all groups and strengthen community connections. Performance occurs in Newark, Delaware, with a period of performance from September 1, 2026, through August 31, 2029.
Mod # | Description | ReasonForModification | Federal Obligation | Date |
|---|---|---|---|---|
| Not listed | $597.5k | 7/23/26 |