Efficient Parallelization Layouts for Large-Scale Distributed Model Training

1 code implementation • 9 Nov 2023 • Johannes Hagemann, Samuel Weinbach, Konstantin Dobler, Maximilian Schall, Gerard de Melo

In this work, we conduct a comprehensive ablation study of possible training configurations for large language models.

Less Is More: A Comparison of Active Learning Strategies for 3D Medical Image Segmentation

1 code implementation • 2 Jul 2022 • Josafat-Mattias Burmeister, Marcel Fernandez Rosas, Johannes Hagemann, Jonas Kordt, Jasper Blum, Simon Shabo, Benjamin Bergner, Christoph Lippert

Since labeling medical image data is a costly and labor-intensive process, active learning has gained much popularity in the medical image segmentation domain in recent years.

Active Learning Benchmarking +4

Parallel Statistical Multi-resolution Estimation

no code implementations • 10 Mar 2015 • Jan Lebert, Lutz Künneke, Johannes Hagemann, Stephan C. Kramer

We show that statistical multi-resolution estimation can enhance the resolution improvement of the plain SOFI algorithm just as the Fourier-reweighting of SOFI.

Super-Resolution