You are here

Reading Seminar on Mathematics of Machine Learning

Recent advancements of machine learning methods lave led to breakthroughs in a wide range of applications. In spite of their empirical successes, the theoretical understanding of machine learning techniques are far from complete. New tools from mathematics and statistics have been showing their power in explaining the mystery and more will be emerging. The purpose of this reading seminar is threefold: First to discuss recent research works that lie in the interface of machine learning, mathematics, statistics and etc; second to expose new research questions and foster collaborations between different groups and even different departments; and third to create new research opportunities for junior researchers and graduate students. All faculty, VAPs, graduate students are welcome to join.

Date Speaker Title
Fri. 2/17
Zhenjie Ren
Université Paris-Dauphine

Regularized Mean Field Optimization with Application to Neural Networks

Fri. 2/24
Paul Hand
Northeastern University

Signal Recovery with Generative Priors

Fri. 3/3
Bryn Elesedy

Symmetry and Generalisation in Machine Learning

Fri. 3/10
Robin Walters
Northeastern University

The Strengths and Limitations of Equivariant Neural Networks

Fri. 3/24
Ziang Chen
Duke University

On Representing (Mixed-Integer) Linear Programs by Graph Neural Networks

Fri. 3/31
Lénaïc Chizat

A Tractable Barycenter for Probability Measures in Machine Learning