I am currently a postdoctoral researcher at the Italian Institute of Technology (IIT) in the Computational Statistics and Machine Learning group directed by Massimiliano Pontil. I defended my PhD in Physics at the Ecole Normale Supérieure de Lyon (ENS Lyon) on October 11th 2016, under the supervision of Patrice Abry and Nelly Pustelnik.
My areas of interest include inverse problems in signal and image processing, model selection techniques, on-the-fly non-smooth convex minimization and scale invariance analysis.
Sparse support vector machine method adapted to deal with imbalanced class sizes.
Download: toolbox_sparseSVM.zip (Matlab. Version: 03-March-2017).
Branch and bound algorithm to jointly estimate the scale invariance properties of bivariate data
This contribution proposes an operational strategy that combines hierarchical Bayesian and l1-Potts/l2-Potts formulations, with the double aim of automatically tuning the regularization parameter and of maintaining computational efficiency.
Download: toolbox_PottsSelect.zip (Matlab. Version: 28-July-2017)
In the context of joint change-point detection shared by multivariate data, we have designed an efficient on-the-fly algorithm leading to an approximate solution to the multivariate total variation minimization problem.
Download: toolbox_TV_OnTheFly.zip (Matlab. Version: 03-November-2016)