Studying the K+Σ- photoproduction within an isobar model employing statistical learning methods
Dimitrios Petrellis
Nuclear Physics Institute, Czech Academy of Sciences
Abstrakt:
We investigate the K+Σ- photoproduction in the framework of an isobar model, where the reaction amplitude is constructed in terms of "tree-level" (lowest-order) Feynman diagrams. In this approach, the coupling constants take effective values, which are fitted to experimental data. However, given the large number of prospective resonances in the energy region of interest (1-2 GeV), the fitting procedure becomes problematic. We propose using regularization, a machine learning method that is customarily used in models with a large number of parameters to prevent overfitting the data. Combined with model selection criteria this process effectively leads to an optimal selection of the resonances involved in the reaction mechanism.
Seminář se koná v úterý 3. května ve 14:00
v zasedací místnosti ÚTEF ČVUT, Praha 1, Husova 240/5.
Ing. Bartoloměj Biskup, Ph.D. tajemník semináře |
doc. Ing. Ivan Štekl, CSc. ředitel ÚTEF |
doc. Dr. André Sopczak předseda NPS, ČS IEEE |
NUCLEAR & PLASMA SCIENCES SOCIETY CHAPTER
IEEE Czechoslovakia section
http://www.ieee.cz/en/nps