Fraunhofer-Institut für Techno- und Wirtschaftsmathematik ITWM

Aufgabenbereiche / Kompetenzen

  • Machine Learning
    • Large Scale Machine Learning
    • Deep Learning
    • Ensemble Methods
  • Pattern Recognition
  • Parallelization


  • Janis Keuper and Franz-Josef Pfreundt. 2016. Distributed training of deep neural networks: theoretical and practical limits of parallel scalability. In Proceedings of the Workshop on Machine Learning in High Performance Computing Environments (MLHPC '16). IEEE Press, Piscataway, NJ, USA, 19-26. DOI:
  • Keuper, Janis; Pfreundt, Franz-Josef, Asynchronous Parallel Stochastic Gradient Descent: A Numeric Core for calable Distributed Machine Learning Algorithms,Proceedings of the Workshop on Machine Learning in High-Performance Computing Environments;, isbn 978-1-4503-4006-9.
    publisher ACM
  • Keuper, Janis; Balancing the Communication Load of Asynchronously Parallelized Machine Learning Algorithms; Journal CoRR, volume abs/1510.01155, Year 2015