Mathematical data science
Gabor Lugosi (University of Pompeu Fabra Barcelona)
Peter Bartlett (University of California Berkeley and Google Research Australia)
Subhro Ghosh (National University of Singapore)
Tselil Schramm (Stanford University)
Nikita Zhivotovskiy (University of California Berkeley)
Data science is a rapidly developing area with tremendous impact on everyday life. Its focus is on information extraction from large amounts of data, usually 'living' in a high dimensional space. The study of the most fundamental problems in Data Science, namely, when and why information extraction is possible, and how to make it efficient, leads to fascinating mathematical problems that require the development of cutting-edge theory.
Like much of Mathematical Data Science, the main theme of this session is on the important role that randomness in high dimension plays. The fact that randomness can be used to expose hidden structures in high-dimensional objects is of crucial importance in modern mathematics (e.g., Asymptotic Geometric Analysis; Harmonic Analysis; Combinatorics; etc.), and Data Science is no exception.