Project Grant N000141712569

This three-year, $240,000 Project Grant from the National Science Foundation's Division of Electrical, Communications and Cyber Systems, under the Engineering program (CFDA 47.041), will support collaborative research between Columbia University and the University of Pennsylvania on scalable and communication-efficient learning-based distributed control. The researchers will develop a foundational and integrated theory of distributed learning-enabled control and approximated distributed...

This $474,000 Project Grant awarded by the National Science Foundation (NSF) under the Computer and Information Science and Engineering (CFDA 47.070) program supports research to develop a suite of neural bandit learning algorithms that leverage recent advances in deep learning theory for efficient neural network model training with incomplete feedback. The key objectives are to: 1) Advance bandit learning methods in more complex neural network architectures and explore new deep learning...

This Project Grant from the National Science Foundation's Computer and Information Science and Engineering program totaling $614,843 will support the development of novel deep learning methods for decentralized and dynamic environments from October 1, 2022 to September 30, 2025. Purdue University will receive the award to advance knowledge on both foundational and practical aspects of local and forward-oriented destructive learning. Specifically, the university will generalize iterative...

This $600,000 Project Grant award from the National Science Foundation (NSF) Computer and Information Science and Engineering (CISE) program supports research by the University of Illinois to develop a theoretical control framework for understanding and improving diffusion-based generative machine learning models. The project aims to establish connections between generative modeling, optimal control theory, and partial/stochastic differential equations. Key technical objectives include enhancing...

This three-year $240,000 Project Grant from the National Science Foundation's Division of Electrical, Communications and Cyber Systems, under the Engineering program (CFDA 47.041), will support research to advance the practical application of distributed learning-enabled control systems. The University of Pennsylvania and Columbia University will collaborate on developing foundational theory and integrated approaches for scalable and communication-efficient distributed control. This includes...

This NSF CISE program Project Grant award of $544,114 to New York University (NYU) is funding research on "NUMERICALLY EFFICIENT REINFORCEMENT LEARNING FOR CONSTRAINED SYSTEMS WITH SUPER-LINEAR CONVERGENCE (NERL)". The project aims to develop new reinforcement learning algorithms that can more efficiently create behaviors for real-world robotic applications, while ensuring operational safety. The research will explore ways to improve learning efficacy and guarantee safety, and will...

This $1.7 million National Science Foundation project grant supports research at West Virginia University and the University of Arkansas, Fayetteville to develop unsupervised continual learning algorithms inspired by neuroplasticity mechanisms observed in electric fish. Funded through NSF's Engineering Directorate (ENG) under the Established Program to Stimulate Competitive Research and Emerging Frontiers in Research and Innovation Brain-Inspired Dynamics for Engineering Energy-Efficient...

This $400,000 Project Grant award from the National Science Foundation (NSF) Computer and Information Science and Engineering (CISE) program (CFDA 47.070) aims to develop a principled mathematical framework for deep learning models that can effectively leverage low-dimensional data structures. The project, titled "COLLABORATIVE RESEARCH: RI: MEDIUM: PRINCIPLED APPROACHES TO DEEP LEARNING FOR LOW-DIMENSIONAL STRUCTURES", will design "white-box" deep neural networks optimized...

This four-year $800,000 Project Grant from the National Science Foundation's (NSF) Computer and Information Science and Engineering program (CFDA 47.070) will support research at the University of California, Los Angeles (UCLA) to develop theoretical tools for understanding deep neural networks (DNNs). The research aims to identify common themes in how artificial and biological systems like the human brain learn. It will investigate the hypothesis that DNNs succeed when learning tasks exhibit...

This $1.2 million project grant from the National Science Foundation's Computer and Information Science and Engineering program will support research and education activities at the University of Washington from October 2022 to September 2026. The project aims to develop a mathematical foundation for deep reinforcement learning by investigating guarantees achievable by neural network policies under different problem structures. Researchers will leverage tools from approximation theory, control...