Skip to main content Skip to main navigation

Project

MetaDL

AI Lab for Metaprogramming in Deep Learning

AI Lab for Metaprogramming in Deep Learning

  • Duration:
  • Research Topics
    Other

MetaDL is a project funded by the BMBF in the field of "Artificial Intelligence". As part of the MetaDL project, we want to generate code for AI applications on a variety of accelerator-based systems based on the AnyDSL framework. Particularly interesting are systems with special instructions for deep learning such as the tensor cores on NVIDIA graphics processors (GPUs) or dedicated AI hardware such as Google's Tensor Processing Units (TPUs). The usefulness of the concepts developed is demonstrated by applications in bioinformatics (analysis of DNA sequences) and image synthesis (noise suppression).

Partners

  • Universität des Saarlandes (UdS: Konsortialleitung)
  • Deutsches Forschungszentrum für Künstliche Intelligenz (DFKI)
  • Johannes Gutenberg Universität Mainz (JGU)

Sponsors

BMBF - Federal Ministry of Education and Research

BMBF - Federal Ministry of Education and Research

Publications about the project

Michael Kenzel; Stefan Lemme; Richard Membarth; Matthias Kurtenacker; Hugo Devillers; Markus Steinberger; Philipp Slusallek

In: Proceedings of the 37th IEEE International Parallel & Distributed Processing Symposium (IPDPS). IEEE International Parallel & Distributed Processing Symposium (IPDPS-2023), May 15-19, St. Petersburg, FL, USA, Pages 736-745, IEEE, 5/2023.

To the publication

Manuela Schuler; Richard Membarth; Philipp Slusallek

In: David Kaeli (Hrsg.). ACM Transactions on Architecture and Code Optimization (TACO), Vol. 20, No. 1, Pages 17:1-17:25, ACM, 12/2022.

To the publication

André Müller; Bertil Schmidt; Richard Membarth; Roland Leißa; Sebastian Hack

In: Proceedings of the 36th ACM International Conference on Supercomputing (ICS). International Conference on Supercomputing (ICS-2022), June 27-30, Pages 20:1-20:11, ACM, 6/2022.

To the publication