Madeleine Udell

Basics
a picture of Madeleine
Photo by Diana Mellon

I am a PhD candidate in Computational & Mathematical Engineering at Stanford University, working with Professor Stephen Boyd. I am interested in modeling and solving large-scale optimization problems, and in finding and exploiting structure in high dimensional data. My methodological interests are driven by the framework of convex optimization and of graph theory, which provide powerful tools for formalizing objectives in statistics and machine learning.

I am currently working on a fast way to maximize a sum of sigmoidal functions , inspired by the venerable spatial theory of voting from political science. My other favorite problems to ponder include how to preserve the information present in heterogeneous data (consisting of graphs and categorical variables as well as numerical data) when squashing it into a euclidean vector space, ways to use graph partitioning to improve the convergence of distributed optimization algorithms, and whether it's possible to speed up bandit learning by using ideas from matrix completion.

In my extracurricular life, I am also a classical harpist, a long-distance runner, and an intrepid cook.


News and Links


Contact