Project Grant 2409702

This $250,000 Project Grant, awarded by the National Science Foundation (NSF) under the Engineering program (CFDA 47.041), supports the development of scalable, robust, and distributed nonconvex approaches for structured tensor recovery. The project, led by Iowa State University, aims to advance the field of tensor analysis to address key challenges in modern data science across various domains, including signal processing, biomedical imaging, machine learning, and quantum information science. The research will focus on exploiting low-dimensional tensor representations to enable efficient tensor recovery from incomplete measurements, with the goal of overcoming computational challenges for large-scale tensors. The project will develop computationally and statistically efficient optimization methods that directly optimize over the low-dimensional tensor structures, as well as distributed optimization techniques that offer similar convergence and recovery guarantees, along with consensus guarantees, for scenarios where privacy or computational resources are a concern. This award will be active from September 1, 2024, through August 31, 2027.