This $200,000 Project Grant award from the National Science Foundation (NSF) Division of Computing and Communication Foundations, under CFDA 47.070 "Computer and Information Science and Engineering", funds research to develop fundamental theory and performance bounds for using machine learning approaches to enable more efficient data compression for applications like wireless communications and mobile devices.
The key products and services to be delivered include: 1) Investigating the factors that influence the convergence of machine learning for lossless and lossy data compression, including understanding when simple models are sufficient; 2) Exploring the use of active learning frameworks to adaptively determine the optimal amount of training data needed; and 3) Deriving performance bounds for learning-based source coding, particularly for sources with memory. The research aims to lead to the development of more powerful data compression algorithms to enable better utilization of wireless spectrum and reduced energy use in mobile devices. The grant was awarded to the Texas A&M Engineering Experiment Station (Tees), a division of the Texas A&M University System, to conduct this 3-year collaborative research project.