Human-Centric Machine Learning: Transparency, Explainability, and Guidance by Human Feedback

Many of the machine learning models used today exhibit black-box behavior: human users cannot easily interpret the internal mechanisms and decision-making processes. And although these models have been trained with huge amounts of data, they rarely take into account the specific needs of users.
This talk showcases methodological advances in human-centered machine learning to control model behavior, reduce bias, and improve alignment with user requirements.

Speaker: Matthias Zeppelzauer, University St. Pölten

This talk is part of the public IfI Colloquium series: https://www.ifi.uzh.ch/en/studies/phd/colloquium/spring-2026.html#Zeppelzauer

You may also like the following events from University of Zurich, Department of Informatics:

• Next Thursday, 19th February, 05:15 pm, From Monologue to Dialogue: Remaking Public Service Communication with Celebrity-as-a-Service in Zurich
• Next month, 19th March, 05:15 pm, Cognitive Scaffolding in the Age of AI: Design Principles for Appropriate Reliance in Zurich