Trusted AI beyond Buzzwordiness: Interdisciplinary Approaches to Fairness, Effective Oversight, and Machine Ethics as Personalized Value Alignment

The talk starts at 13:00, coffee will be served from 12:45. Please help us cut down waste by bringing your own mug.

Abstract: In this talk, Kevin will provide an overview of his interdisciplinary work at the intersection of explainable AI (XAI), fairness monitoring, and machine ethics, particularly in the context of effective human oversight (as required, for instance, by the EU AI Act). He will share insights into recent work on a runtime monitor for individual fairness, a neuro-symbolic approach combining formal methods and safe reinforcement learning for machine ethics and personalized value alignment, and interdisciplinary efforts to better understand calibrated trust. He will also briefly introduce the Center of European Research in Trusted AI (CERTAIN).