Modern data, even though large and high-dimensional, often contains a “simple” structure, containing only a few degrees of freedom. Leveraging this structure allows for the design of efficient algorithms to collect, process, and communicate the data. For example, a large but low-rank matrix can be completed after observing a relatively few of its entries at random. Or, in molecular imaging, the locations of a small number of point sources can be resolved from a blurry image. This Interest Group is dedicated to the study of similar problems in order to improve the state of the art, and is also designed to facilitate collaboration between its members.
In addition to theoretical underpinnings, we also strive to develop fast and scalable algorithms for such problems, borrowing tools from areas such as: stochastic and distributed optimization, compressed sensing, (numerical) linear algebra, (high-dimensional) statistics, approximation theory, probability and others.
Part of The Alan Turing Institute-Lloyd’s Register Foundation Programme for Data-Centric Engineering