Context and goal

Big data raises many tough issues, including: storage, transferring, processing, and interpreting. These issues are major challenges of the upcoming decades. This evolution is mirrored both in mathematical optimization and machine learning research by the increasing number of large datasets now available to researchers, along with the number of benchmarks and challenges associated to these datasets exhibiting large scales in all the dimensions of learning problems: the number of examples, the number of features, the number of tasks, and the number of models.

The goal of Titan is to form an alliance between researchers resp. from mathematical optimization and machine learning to tackle these big data challenges in general, with a special interest in processing in astrophysics data. One of our main our research is to develop stochastic and distributed learning and optimization methods for very large antenna networks in radioastronomy.

Scientific Events of 2015

Contributions, realisations of 2015

This year's activities are detailed in the presentation given at the Mastodons workshop in december 2105.

People who joined the team during this year 2015: Laurent Condat (Gipsa-Lab), Jalal Fadili (Univ. Caen)

Initial project members

Previous mastodons projects (with related events and highlights)

The project Titan is the merge of two brothers Mastodons projects of previous years: and Display led by Andre Ferrari, Nice and Gargantua (led Zaid Harchaoui, previously with LJK Grenoble, and now at NYU). This fusion was a clear success : we have had many interactions and pertinent cross-influence. See precise elements on the presentation given at the Mastodons workshop.