We are a research group at UC Berkeley that focuses on mathematical optimization and its applications to natural language processing, machine learning and deep learning.
On November 15th Alexandre Gramfort, Researcher at INRIA, will give a talk entitled “From Gap safe rules to working sets for faster sparse solvers”. This will take place at 10am in Soda Hall, room 405.
Invited talk "Stochastic Gradient Descent: Going As Fast As Possible But Not Faster" by Alice Schoenauer-Sebag
On Friday 29th of September we will have the visit of Alice Schoenauer-Sebag, who will give a presentation on her recent paper Stochastic Gradient Descent: Going As Fast As Possible But Not Faster.
The paper “Breaking the Nonsmooth Barrier: A Scalable Parallel Method for Composite Optimization”, by Fabian Pedregosa, Rémi Leblond and Simon Lacoste-Julien gets a spotlight at the prestigious NIPS conference, which will be held in Los Angeles in December. This work will also be presented in the upcoming baylearn workshop.