The work of CP researchers on gender bias in neural Information Retrieval models in news  

Decisions made by learning-based systems can be discriminatory or racist, caused by the underlying biases in data and models. Navid Rekab-Saz and Markus Schedl from the Computational Perception Institute investigate this topic in the context of search engines, and find out that use deep learning models are particularly gender biased. Read the full story here:

NEWS 08.07.2020