Deep Learning Compute Cluster (partial financing)

• Subject Area: Computer Science
• Term: Funded in 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 539134284

The Freiburg Deep Learning Compute Cluster 2.0 extends and renews the existing Freiburg Deep Learning Compute Cluster with modern, powerful and energy-efficient GPU and CPU compute units, as well as storage capacities, to explore new transformative deep learning methods. The new methods to be developed using the requested compute cluster support a variety of research topics, including generative language models, multimodal models with language and images, autonomous driving, robotics, bioinformatics, medical imaging and medicine in general, energy-efficient machine learning methods, algorithmic fairness, climate science, and automated machine learning. The compute cluster brings efficient and energy-saving machine learning to the forefront, laying a foundation for green digitization.

• DFG Programme: Major Research Instrumentation
• Major Instrumentation: Deep Learning Rechencluster (Teilfinanzierung)
• Instrumentation Group: 7000 Datenverarbeitungsanlagen, zentrale Rechenanlagen
• Applicant Institution: Albert-Ludwigs-Universität Freiburg
• Leader: Professor Dr. Frank Hutter, Ph.D.