Efficient Optimization Algorithms for Interpretable Machine Learning

Behind all supervised learning problems is an optimization problem. Solving these problems reliably and efficiently is a key step in any machine learning pipeline. This talk looks at efficient optimization algorithms for a variety of machine learning problems (in particular, sparse learning problems). In the first part of the talk, we look at efficient algorithms for constructing knockoff features for statistical inference. In the second part, we will look at l0-penalized and constrained optimization problems and a class of efficient algorithms for training these non-convex problems while providing guarantees on the quality of the solution.