In the context of IRISA D5 (Signal, Image & Robotics) department’s scientific seminars, we held a two-day workshop on “Sparse Models and Machine Learning” , on October 15-16, 2012.
Sparse representations and models have become a central concept in the field of signal processing and data modeling, and are gradually building bridges with the area of machine learning. In parallel, these concepts are becoming operative in a number of apllicative fields, such as brain imaging, distributed communications, multimedia, etc…
This two-day workshop offered an overview of the state of the art of this stimulating field and discussions of very exciting focal points, where sparse representations, machine learning and applications are converging to new fundamental and pratical paradigms.
The worhshop was opened ta academy and industry spacialists, but also to PhD students, post-doc and colleagues from neighboring domains, who want to deepen their understanding od the field and its current theoretical and pratical challenges. The program has been labeled « complément scientifique » by Ecole Doctorale Matisse.
You can download here the speakers’ slideshows :
- Rémi Gribonval: Inverse problems and sparse models jap_gribonval1 ,jap_gribonval2
- Francis Bach: Stuctured sparsity and convex optimization jap_bach1_2,
- Pierre Vandergheynst: Emerging applications of sparse representationsjap_vandergheynst
- Enrico Magli: Compressed sensing for distributed communications ,jap_magli
- Alexandre Gramfort: Sparse methods for brain imaging ,jap_gramfort
