Münster 2017 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

T: Fachverband Teilchenphysik

T 21: Experimentelle Methoden 1 (Computing, Machine Learning, Statistik)

T 21.4: Vortrag

Montag, 27. März 2017, 17:30–17:45, JUR 253

Application of the VISPA web-platform for deep-learning based physics analyses — •Benjamin Fischer, Martin Erdmann, Robert Fischer, Erik Geiser, Christian Glaser, Gero Müller, Thorben Quast, Marcel Rieger, Martin Urban, Florian von Cube, David Walz, and Christoph Welling — Physics Institute III A, RWTH Aachen

VISPA (Visual Physics Analysis) is a web-platform allowing to conduct browser-based analyses including access to facilities for exploiting deep learning methods. Opportunistic resource allocation and modular extendability ensure the versatility to tackle a broad range of challenges.

VISPA's architectural concept and basic functions will be briefly outlined alongside the most recent developments. An overview on the available specialized tools for physics analyses will be given and explicitly illustrated in the scope of an in-progress analysis using deep learning. These include native-format data browsers for various file formats (e.g. ROOT) and methods for exploring highly ramified file-trees.

VISPA is used for high energy physics and astro particle physics as well as for university-level education.