Dear all / Bonjour à tous,

We are happy to invite you to Jose Gallego-Posada's PhD defense on August 15th at 9h30 am (hybrid mode).

Vous êtes cordialement invité.e.s à la soutenance de Jose Gallego-Posada,

le mardi 15 aout à 9h30 am (mode hybride).

Jury

Abstract

This thesis presents a holistic perspective on the use of constrained
optimization as a framework for inducing complex behaviors in modern
machine learning models. We introduce fundamental concepts and algorithms
from constrained optimization through a machine learning lens. We present
several case studies highlighting the advantages of the constrained
approach, algorithmic innovations for solving constrained problems
involving neural networks, and an open-source library that aims to
facilitate the adoption of the proposed constrained 

framework by the machine learning community.