Discrimination, fairness, and algorithms
Seminar series (March–April 2025)
Johannes Gutenberg University Mainz
Registration required for Informatik students who intend to receive credit, via Jogustine (course code 08.079.24002).
Registration appreciated for everyone, bottom of page
Outline
Predictive models and decision-making algorithms increasingly influence critical areas such as hiring, lending,
healthcare, and criminal justice. Significant concerns have emerged about how biases in data, models, and outputs can lead to unfair outcomes, including discrimination based on protected attributes such as gender or ethnicity.
Fair machine learning (or algorithmic fairness) refers to the study and practice of ensuring that prediction-based decision-making processes, particularly those driven by machine learning and artificial intelligence, operate in manners that are fair, equitable, and just. This field seeks to develop methods for detecting, mitigating, and preventing unfair or inequitable bias, ensuring that algorithmic systems treat all individuals and groups fairly while balancing considerations of accuracy, efficiency, and ethical standards.
Despite promising developments, the path towards fair AI systems has proven difficult to traverse for many
reasons. Difficult tradeoffs must be addressed during model development, and bias mitigation interventions can have unexpected effects. In addition, numerous organizational and regulatory challenges can become obstacles to achieving fairness in practice. The seminar series “Discrimination, fairness, and algorithms” focuses on timely moral, legal, technical challenges that must be addressed in the domain of fair machine learning broadly construed.
The seminar seeks to foster a deeper understanding of these challenges and explores interdisciplinary approaches to building fairer and more ethical AI systems. The seminar series features invited speakers from a diverse range of disciplines, including computer science, data science, philosophy, and law. Each session explores a critical challenge in relation to fairness in prediction-based decision-making and offers fresh perspectives and actionable insights into how fairness can be embedded into the fabric of AI development and deployment. The series is open to faculty, students, and researchers. Attendees will have the opportunity to engage with the invited speakers and contribute to the seminar’s discussion. Join us to be part of ongoing dialogue on how we can collectively work towards building more equitable AI for all.
Schedule
All lectures will be held at 14:00 in N6, NatFak (directions)
10.3. Hilde Weerts (University of Eindhoven); Raphaele Xenidis (SciencesPo)
21.3. Michele Loi (AlgorithmWatch); Otto Sahlgren (Tampere University)
28.3 Mattia Cerrato (JGU Mainz); Otto Sahlgren (Tampere University)
11.4 Corinna Hertweck (University of Zurich); Jose Alvarez (KU Leuven)
25.4 Prof. Dr. Stefan Kramer (JGU Mainz)
Entschuldigung, aber du musst angemeldet sein, um dieses Formular sehen zu können.