Machine learning algorithms may become biased because of the data they are trained on or the way they are coded, thereby delimiting the ethics of technological innovation. On the one hand, bias represents a key challenge to the development of responsible AI. On the other hand, identifying and revaluating bias in data and code might offer technological solutions to social problems. In this session, we discuss the challenges and potentials of working with algorithmic bias.
Program: Introduction: Sine N. Just, Professor, RUC Scientific Talk: “Bias in machine learning: let the one without bias cast the first stone” Christina Lioma, Professor, University of Copenhagen, Dept. Computer Science Business Talk: “Why tech companies should lead in advancing responsible” AI Mikael Ekman, Director of Public Policy, Microsoft Denmark & Iceland Business Talk: “Infrastructures for AI fairness and trust”, Meeri Haataja, CEO & Co-Founder, Saidot Scientific Talk: “Black-boxing the future – algorithmic bias in open-ended systems with long-term externalities”, Alf Rehn, Professor, University of Southern Denmark Debate: “What are the challenges of and opportunities for working with algorithmic bias?”, moderator: Sine N. Just, Professor, RUC
Session coordinator: Sine Nørholm Just, RUC
Facilitator: Sine Nørholm Just, RUC
Algorithmic bias: What is it? How can we work with it?