PhD proposal: Exploring the differential privacy in the context of distributed and collaborative machine learning for industrial IoT

The objective of this thesis is to explore the use of differential privacy with distributed supervised as well as non-supervised ML methods in IoT/multiCloud environment where data are either disseminated or sensitive. Because of the data complexity inherent to the big data framework, different advanced ML methods such as clusterwise regression will be developed in the context of Federated Learning.

The candidate must have a strong background in statistics and machine learning.

The thesis will be co-supervised by:

  • Prof. Samia Bouzefrane (
  • Prof. Pierre Paradinas (
  • Dr Ndeye Niang (

The candidate is expected to start as soon as possible

Ph.D. Position in Differential Privacy and Machine Learning in IoT

Recent publications