Biography

Ulf Schoneberg is Head of Data Science at The unbelievable Machine Company with a mathematician background. Ulf took part at various organizations such as zbMATH, Humboldt University, Loomes and Ericsson as a software architect, data scientist and researcher. His research interests include natural language processing methods for text analysis and neural networks of all kinds.

Talk

Post-Blackbox AI

Machine learning models as deep neural networks trained on big data have been shown to be highly accurate for many applications, even though their complexity virtually makes them black-boxes. We talk about a novel model-agnostic system that explains the behavior of complex models with high-precision rules representing local, “sufficient” conditions for predictions. We show a program to explain which areas of an image the AI has found important for their decision.