Explaining AI outputs has been a topic I have worked on implementing quite a bit. Last May I gave a talk Der KI auf die Finger geschaut (Keeping an eye on the AI) to mathematicians and actuaries at the University of Göttingen.

I framed the methods to analyse neural network outputs in terms of risk attribution which many insurance actuaries are familiar with. Even though this was a presentation-only event, you can probably tell my inclination for hands-on by the examples and the number of links. Without further ado, get the (German) slides.

Workshop: Keeping an eye on the AI

Now that the topic moves more and more mainstream, I have revisited the topic and expanded it into a full-day workshop featuring theory and hands-on included.

• Location: Munich (Garching)
• Date: Monday December 9th, 9-17.30h with breaks
• Bring a computer with PyTorch & Jupyter Notebooks installed.
• The workshop will be in English (unless all participants prefer German).
• Limited to 8 participants, so you get individual attention during the hands-on.
• Price: €600 + VAT.
• more information at MathInf

Register via email.

P.S.: Did you know that you can also book custom inhouse workshops? They have the advantage that not only we cover what matters most to you but you can also bring your own applications.