Title: Large-scale optimization and deep learning techniques for data-driven signal processing
Abstract: With more advanced technology, everyday life has become more immersed in data of various forms such as images, sounds, and time series. Consequently, the computational methods necessary to process and infer information from these data have become paramount. In this work, we develop several algorithms for data-driven signal processing. The first set of algorithms is based on large-scale optimization methods for recovering signals from noisy observations in applications from medical imaging and fluorescence lifetime imaging microscopy. The second set of algorithms is based on deep-learning techniques for problems arising in synthetic aperture radar and genomics.
