Technologies that collect, store and analyse massive amounts of heterogeneous data in near real time have brought AI into many industrial “intelligent systems”: biometrics, vehicles with partial delegation, recommendation engines, virtual assistants, or predictive maintenance of complex infrastructures such as energy networks and aircraft, where detecting weak signals early allows components to be replaced before they fail.
The scientific challenges remain numerous. Although closely intertwined in industrial applications, three scientific axes structure the chair’s activity.
Axis 1 — Complex data
The complexity of the information likely to feed AI systems: data of various types and formats, collected at scale and often in real time.
Axis 2 — Delegating difficult tasks
The difficulty of the tasks we wish to delegate, at least in part, to intelligent and autonomous systems.
Axis 3 — Trustworthy AI
In many cases, the level of delegation granted to these systems will depend on the answers methodological research provides to issues of ethics, reliability and energy frugality, so that AI becomes acceptable and produces decisions that the end user can interpret.
News & opportunities
Workshops, internship offers and PhD positions related to the chair.