The DPG Spring Meeting in Bonn had to be cancelled! Read more ...
Parts | Days | Selection | Search | Updates | Downloads | Help
HK: Fachverband Physik der Hadronen und Kerne
HK 20: Instrumentation III
HK 20.2: Talk
Tuesday, March 31, 2020, 17:15–17:30, J-HS C
Machine learning approaches for multi-neutron detection with NeuLAND — •Jan Mayer and Andreas Zilges for the R3B collaboration — Institute for Nuclear Physics, University of Cologne
Upcoming experiments featuring Reactions with Relativistic Radioactive Beams (R3B) at the Facility for Antiproton and Ion Research (FAIR) require precise multiplicity and energy information of the emitted neutrons. NeuLAND, the New Large Area Neutron Detector, is dedicated to the detection of up to five high-energy neutrons.
Reconstruction of the multiplicity and the first interaction points from the complex, superimposed hit patterns is challenging. As an alternative to classical methods, we study modern machine learning methods ranging from simple scikit-learn classifiers to deep neural networks with keras.
Here we give an overview of challenges, solutions, and results obtained with a diverse set of approaches and ideas for integration in a data analysis pipeline.
Supported by the BMBF (05P19PKFNA) and the GSI (KZILGE1416).