E. Breuninger – Color The Blackbox

June 13th, 5.45 – 6.15 pm

When it comes to explainable AI, understanding pertinent models is a challenge where data transformations are an unmistakably important component. We explored the feasibility of using a trained Kohonen map to capture the underlying patterns within the data as an unsupervised learning dimension reduction technique. This permits the interpretability and transparency of the model while achieving the data transformation goal. Kohonen maps emerge not only as a powerful tool for exploring, understanding and explaining complex data as they preserve the topological structures within the data but also make the AI more trustable.


Dr. Saurabh Steixner-Kumar, Data Scientist | E. Breuninger

E.Breuninger_Dr.-Saurabh-Steixner-Kumar_CFP-squareSaurabh Steixner-Kumar is a Data Scientist at the center of excellence at E. Breuninger. He has an engineering background with a couple of stints with Airbus and a doctorate in neuroscience. After spending several years doing academic research at Max Planck Institute and UKE Hamburg, he now pushes for innovative solutions to data science problems. He is convinced that scientific research is an integral part of any industry in the modern world.