Annonce en lien avec l’Action/le Réseau : aucun
Laboratoire/Entreprise : MAGNET Project-Team, INRIA Lille
Durée : 3 years
Contact : aurelien.bellet@telecom-paristech.fr
Date limite de publication : 2016-06-30
Contexte :
The Inria Magnet team has an opening for a PhD position on privacy-friendly distributed machine learning. The topic of this Ph.D. project is the development of algorithms to perform tasks which are currently done with complete knowledge of the data in a much more distributed and privacy-friendly way. In particular, the problem of learning a model from the distributed data of many users who do not want to share their data will be considered.
Sujet :
The overall goal of this project is to get a better understanding of the potential of private distributed learning, to construct theory supporting this understanding and algorithms exploiting it. Several scenarios will be investigated, depending for instance on whether synchronization between users through a global server is possible. To guarantee privacy, we will study how to adapt methods from secure multi-party computation to statistical machine learning in order to develop efficient algorithms for which one can control the amount of disclosed sensitive information. Where possible, the potential of the obtained results will be demonstrated in real-world applications, such as ride sharing (predicting rides and matching between users without centralization of mobility logs) and crowdsourcing (task recommendation to users without compromising their personal data).
Profil du candidat :
– strong background in machine learning, statistics and algorithms
– additional knowledge on distributed distributed systems and cryptography would be appreciated
– some experience in implementation and experimentation is expected
– good writing and speaking English skills
Side information:
Salary: 1958€ the first two years and 2059€ the third year
Salary after taxes: around 1600€ the 1st two years and 1679,76 € the 3rd year (benefits included).
Possibility of French courses.
If interested, send to Jan Ramon (jan [dot] ramon [at] inria [dot] fr) and Aurélien Bellet (aurelien [dot] bellet [at] inria [dot] fr) as soon as possible: your CV, a statement of interest, and any supporting documents (e.g., recommendation letters) you may consider helpful.
Formation et compétences requises :
The candidate will work under the supervision of Jan Ramon and Aurélien Bellet. The PhD will begin in October 2016. In France, a PhD should be completed at the end of the 3rd year.
Adresse d’emploi :
The working environment is located at the Inria research center in Lille, France. Combining computer sciences with mathematics, Inria’s 3,400 researchers strive to invent the digital technologies of the future. Educated at leading international universities, they creatively integrate basic research with applied research and dedicate themselves to solving real problems, collaborating with the main players in public and private research in France and abroad and transferring the fruits of their work to innovative companies. The researchers at Inria published over 4000 articles a year. They are behind over 270 active patents and 105 start-ups. The 171 project teams are distributed in eight research centers located throughout France. Lille is the capital of the north of France, a metropolis with 1 million inhabitants, with excellent train connection to Brussels (34 min), Paris (1h) and London (1h30), and tons of opportunities for social life.