Post-Doctoral Research Visit F/M Post-doc Position on Fairness and Privacy in Trustworthy machine Learning (F/M)

Eighteen-month postdoctoral position in the MAGNET team at Inria Lille, supervised by Michaël Perrot. The team develops ethically acceptable machine-learning algorithms focused on fairness, privacy and decentralised learning, and on empowering end users of artificial intelligence. Deploying machine learning at scale, for instance in medical diagnosis or automated hiring, requires more than accuracy: models should not be biased against subgroups of the population, and released models should not leak information about their training data.

The position aims to make the team’s fairness- and privacy-preserving approaches more accessible to the machine-learning community. A first objective is to extend and benchmark the FairGrad library, an easy-to-use approach to enforcing fairness, and integrate it with existing fairness libraries (AIF360, Fairlearn) and general-purpose libraries such as scikit-learn. A second objective builds on recent team results showing that it is possible to learn from locally private data with a controlled bias.

Start date 1 March 2027; gross salary €2,788 per month.