WERK1 (Eventspace XL), Atelierstraße 29, 81671 München
Deep neural networks have been successfully applied to many domains, including very sensitive domains like health-care, security, fraudulent transactions and many more. These domains rely heavily on the predictions accuracy of the model and even one overconfident decision can result in a big problem.
In contrast to regular neural networks, Bayesian neural networks learn a probability distribution (usually modelled as a Gaussian) over the networks weights instead of single numeric values for each. In this talk we will explore the motivation, the mathematical underpinnings and how these networks are implemented in practice.
6:00 PM Doors Open at Werk1 Innovation Hub
6:30 PM Welcome and Introduction by Maximilian Gerer, CTO of e-bot7
6:45 PM Bayesian Neural Networks - Neural Networks that know what they don't know #JonSnow by Till Bauer, Data Scientist at e-bot7
7:20 PM Bayesian Updating - Labels that know what they don't know by Dr. Stephan Werner, CEO & Co-Founder of Risklio
8:00 PM Drinks and Networking