The MIT Institute for Foundations of Data Science (MIFODS) is an NSF funded interdisciplinary effort to develop the theoretical foundations of data science through integrated research and training activities. To achieve this goal, MIFODS stimulates research and educational interactions between mathematics, statistics and theoretical computer science, both within MIT and in the research community at large.
The five research themes of MIFODS
Sublinear algorithms, local algorithms and robust statistics
Statistical and Computational Tradeoffs
Learning under complex structure
Non-convex optimization and deep learning
Graphical models, Exchangeable models and Graphons
Meet the MIFODS investigators