• English
  • Deutsch
  • Log In
    Password Login
    Research Outputs
    Fundings & Projects
    Researchers
    Institutes
    Statistics
Repository logo
Fraunhofer-Gesellschaft
  1. Home
  2. Fraunhofer-Gesellschaft
  3. Konferenzschrift
  4. Distributed training of deep neural networks: Theoretical and practical limits of parallel scalability
 
  • Details
  • Full
Options
2016
Conference Paper
Title

Distributed training of deep neural networks: Theoretical and practical limits of parallel scalability

Abstract
This paper presents a theoretical analysis and practical evaluation of the main bottlenecks towards a scalable distributed solution for the training of Deep Neural Networks (DNNs). The presented results show, that the current state of the art approach, using data-parallelized Stochastic Gradient Descent (SGD), is quickly turning into a vastly communication bound problem. In addition, we present simple but fixed theoretic constraints, preventing effective scaling of DNN training beyond only a few dozen nodes. This leads to poor scalability of DNN training in most practical scenarios.
Author(s)
Keuper, J.
Fraunhofer-Institut für Techno- und Wirtschaftsmathematik ITWM  
Preundt, F.-J.
Fraunhofer-Institut für Techno- und Wirtschaftsmathematik ITWM  
Mainwork
2nd Workshop on Machine Learning in HPC Environments, MLHPC 2016  
Conference
Workshop on Machine Learning in HPC Environments (MLHPC) 2016  
Supercomputing Conference & Expo (SC) 2016  
Open Access
DOI
10.1109/MLHPC.2016.006
Language
English
Fraunhofer-Institut für Techno- und Wirtschaftsmathematik ITWM  
  • Cookie settings
  • Imprint
  • Privacy policy
  • Api
  • Contact
© 2024