Project Grant 2343599

This $300,000 Project Grant award from the National Science Foundation's Computer and Information Science and Engineering (CFDA 47.070) program supports collaborative research on the mathematical and algorithmic foundations of multi-task reinforcement learning. The research aims to address the challenge of data efficiency in reinforcement learning, developing new approaches that can learn multiple related tasks simultaneously using less data and computational resources compared to learning each task individually. Key products of the project will include novel learning algorithms, theoretical analyses, and experimental evaluations through multi-robot navigation simulations. The lead institution is Virginia Polytechnic Institute & State University (Virginia Tech), leveraging the university's expertise in areas like artificial intelligence, robotics, and data analytics. The project period runs from April 1, 2024 through March 31, 2027.