Dienstag, 08.03.2016, 13:00-19:00 Uhr im H1510, Helmholtzbau
Leitung des Tutorials
Dr. Hans-Georg Zimmermann, Siemens AG
Dr. Ralph Grothmann, Siemens AG
Inhalt des Tutorials
The predictive analytics is a race between the increasing complexity of the real world and our accelerating ability to mathematically represent it by means of information-technology-related capabilities, such as neural network models.
From a mathematical point of view, neural networks allow the construction of models, which are able to handle high-dimensional problems along with a high degree of nonlinearity. Our philosophy is beyond purely data-driven modeling: The application of neural networks should be based on a deep understanding of the underlying mathematics, first principles on dynamical systems as well as prior (economic) domain knowledge. The tutorial will introduce feedforward neural networks, including deep neural networks, for non-linear regression tasks and time-delay recurrent neural networks for modeling dynamical systems. Examples from real-world real-world industrial applications will be given that outline the merits of such a modeling approach. Among others we will deal with the modeling of e.g. the energy supply from renewable sources, energy load forecasting as well as the forecasting of commodity prices and the identification of features responsible for component failures.
Zielgruppe des Tutorials
Interessierte an nichtlinerarer Zeitreihenanalyse und Klassifikation.