Mathe im Beruf Detailseite

Vortragende Person/Vortragende Personen:
Dr. Rebekka Burkholz

Abstract:

Deep learning continues to achieve impressive break-throughs across disciplines but relies on increasingly large neural network models that are trained on massive data sets. Their development inflicts costs that are only affordable by a few labs and prevent global participation in the creation of related technologies. In this talk, we will ask the question if it really has to be like this and discuss some of the major challenges that limit the success of deep learning on smaller scales, which we often face in case of socioeconomic problems. By casting a neural net-work as a cascade process that evolves on a complex network, we propose to tackle major computational chal-lenges that are induced by the overparameterization of modern neural network models and discuss advantages of this view on modelling systemic risk inherent in international food trade.