INFORMATIK 2022

Trustworthy AI in Science and Society [EN]
29.09, 14:30–18:00 (Europe/Berlin), ESA West 120
Sprache: English

Chairs:
Marian Margraf, Fraunhofer AISEC
Gerhard Wunder, FU Berlin
Eirini Ntoutsi, FU Berlin
Maximilian Poretschkin, Fraunhofer IAIS
Franziska Boenisch, Fraunhofer AISEC
Karla Markert, Fraunhofer AISEC

Artificial intelligence (AI) has made its way into a broad variety of sensitive applications, such as health care, hiring processes, and autonomous service. Thereby, it has a direct impact on our daily lives and potential malfunctioning could cause severe damage for individuals and society. Therefore, the topic of trustworthiness in AI has moved into focus. With this workshop we aim to cover different perspectives of trustworthy AI, from technical to societal including topics on security, fairness, transparency, explainability, safety, and privacy. How can we technically evaluate the distinct aspects of trustworthiness? How do they interfere with one another and how can we improve them? How can methods to implement trustworthiness be applied in practice by a broad spectrum of users and applications, and how do we make sure to eliminate risks? What does it take to include and educate non-technical users on AI trustworthiness and how can society benefit from these insights? Finally, what is needed to create trustworthiness in AI?

Further Information: https://www.zvki.de/


14:30 Session 1

The Influence of Training Parameters on Neural Networks' Vulnerability to Membership Inference Attacks
Oussama Bouanani and Franziska Boenisch

The integrative model of organizational trust from Mayer et al (1995). Actually a great theory that stays perfectly valid for Human-AI Collaboration research
Lilian Tai Do Khac and Michael Leyer

GAFAI: Proposal of a Generalized Audit Framework for AI
Thora Markert, Fabian Langer and Vasilios Danos

15:30 Kaffeepause

16:00 Session 2

Keynote
Vince Istvan Madai

Measuring Gender Bias in German Language Generation
Angelie Kraft, Hans-Peter Zorn, Pascal Fecht, Judith Simon, Chris Biemann and Ricardo Usbeck

Fairness in Regression - Analysing a Job Candidates Ranking System
Karla Markert, Afrae Ahouzi and Pascal Debus