Mathematical Programming Group develops theory, methods, and software for efficiently solving optimization and decision problems using modern computing systems.
The group focuses on large-scale optimization problems and their applications in machine learning and data science. These problems are often characterized by vast amounts of data potentially distributed over an extensive network with restricted access, a large number of parameters, and complex problem models involving non-smooth and non-convex loss functions.
Our primary research topics include:
Disciplined convex programming
Stochastic optimization and variance reduction methods
Distributed and federated optimization
Monotone operator theory and variational inference
Implicit regularization and theory of deep learning