Guillaume Obozinski

As of January 2019, I joined the Swiss Data Science Center (SDSC) as Deputy Chief Data Scientist. I am based at the office of the SDSC located on the EPFL campus in Lausanne.

I was research faculty at Ecole des Ponts - ParisTech from 2013 to 2018, in the Imagine team of the Laboratoire d'Informatique Gaspard Monge.

My area of research is Machine Learning and its applications.

News

Posted on January 11th, 2017

- As of January 2019, I joined the Swiss Data Science Center as Deputy Chief Data Scientist. I am based at the office of the SDSC located on the EPFL campus in Lausanne.

Posted on June 26th, 2017

- I just posted online the slides of the short course on Probabilistic Graphical Models that I am giving at the INIT/AERFAI Summer School in Machine Learning in Benicassim, June 26-30 2017.

Posted on January 20th, 2016

- On February 3rd 2016, Ecole des Ponts is organising a colloquium entitled
Perspectives and New Challenges in Data Science . Follow the link for more details.

Posted on September 3rd, 2014

- We a currenty looking to hire a PhD student with Nikos Komodakis and Olivier Tournaire for a computer vision project on Large Scale semantization of 3D urban mesh models. This position has been filled.

Posted on June 1st, 2014

- With Laurent Najman from ESIEE Paris, we are organizing the colloquium of the Labex Bézout on Data Science and Massive Data Analysis on June 12th 2014 in Marne-la-Vallée.

Posted on May 20th, 2014

- With Nabil Rachdi from Airbus group, we are organizing a workshop on statistical learning methods for computer experiments on May 28th, 2014.

Posted on March 28, 2012

- I am coorganizing an ICML 2012 workshop Sparsity, Dictionaries and Projections in Machine Learning and Signal Processing, taking place on June 30th 2012.

- I will be teaching a course on Probabilistic Graphical Models for Information retrieval at the Russian Summer School in Information Retrieval (RuSSIR 2012), August 6-10, 2012, in Yaroslavl.

Posted on December 2, 2011

- I am coorganizing the NIPS workshop Sparse Representation and Low-rank Approximation
- Our paper with Edouard Grave was accepted at NIPS:
- Trace Lasso: a trace norm regularization for correlated designs E. Grave, G. Obozinski, F. Bach, Advances in Neural Information Processing Systems (NIPS), 2011. [arXiv technical report]
- I will be presenting recent work on Relaxation of Combinatorial penalties in the Workshop on Discrete Optimization in Machine Learning (DISCML) 2011