BAT 2.0: Development and applications

• Applicant: Professor Dr. Kevin Kröninger
• Subject Area: Nuclear and Elementary Particle Physics, Quantum Mechanics, Relativity, Fields
• Term: from 2018 to 2023
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 400959267

• Final Report Year: 2023

Final Report Abstract

Statistical data analysis is a central tool in all areas of scientific research. Bayes' theorem allows a consistent formulation of questions which are typically asked in data analysis, especially for the determination of parameters and the comparison of models. However, the evaluation of the corresponding key variables, e.g., the a posteriori probabilities and the evidence, is often difficult, especially for complex models with many parameters, as they are often found in science. The latter are usually only accessible with custom-built software tools. The increasing amount of data generated in scientific experiments must be accompanied by the development of tools that can process this data. The first goal of the project is to develop a robust tool for Bayesian inference, BAT.jl, which can be used in different scientific disciplines. This tool must allow fast and reliable computation of statistical quantities for high-dimensional and complex models. At the same time, it must provide a platform for the development, comparison and application of different algorithms, which guarantees comparable and controlled conditions. Furthermore, the tool must be platform-independent and not focused on the specific requirements of a particular field of work. The second goal of the project is to apply this tool to a specific class of problems, namely parameter estimation of complex physics models, in order to show the full potential of the software and to highlight case studies. For this purpose, a tool derived from BAT.jl has to be further developed, the EFTfitter.jl, which can be used to fit models to data. This application is located at the interface between scientific modeling, statistical inference, and numerical methods. It is also intended to act as a representative of a class of problems often encountered in science, thus highlighting the interdisciplinary nature of the project. The third goal of the project is to apply EFTfitter.jl to two concrete problems in the field of particle physics: The first problem is to estimate the free parameters in models of effective field theory, which additionally contains contributions from operators of dimension six. The model will be compared with a variety of experimental data from the field of top quark and bottom quark physics. The second problem is the determination of the free parameters of Monte Carlo generators (tuning). Here, the Herwig event generator was chosen and a comparatively large selection of observables was used.

Publications

• BAT.jl Upgrading the Bayesian Analysis Toolkit. EPJ Web of Conferences, 245(2020), 06001.
  Caldwell, Allen; Grunwald, Cornelius; Hafych, Vasyl; Kröninger, Kevin; La, Cagnina Salvatore; Schulz, Oliver & Shtembari, Lolian
  
• Constraining top-quark couplings combining top-quark and $$\varvec{B}$$ decay observables. The European Physical Journal C, 80(2).
  Bißmann, Stefan; Erdmann, Johannes; Grunwald, Cornelius; Hiller, Gudrun & Kröninger, Kevin
  
• Correlating uncertainties in global analyses within standard model EFT matters. Physical Review D, 102(11).
  Bißmann, Stefan; Erdmann, Johannes; Grunwald, Cornelius; Hiller, Gudrun & Kröninger, Kevin
  
• Study of interference effects in the search for flavour-changing neutral current interactions involving the top quark and a photon or a Z boson at the LHC. The European Physical Journal Plus, 135(3).
  Barros, Maura; Castro, Nuno Filipe; Erdmann, Johannes; Geßner, Gregor; Kröninger, Kevin; La, Cagnina Salvatore & Peixoto, Ana
  
• BAT.jl: A Julia-Based Tool for Bayesian Inference. SN Computer Science, 2(3).
  Schulz, Oliver; Beaujean, Frederik; Caldwell, Allen; Grunwald, Cornelius; Hafych, Vasyl; Kröninger, Kevin; Cagnina, Salvatore La; Röhrig, Lars & Shtembari, Lolian
  
• Enhancing fits of SMEFT Wilson coefficients in the top-quark sector. Proceedings of 40th International Conference on High Energy physics — PoS(ICHEP2020) (2021, 2, 26), 323. American Geophysical Union (AGU).
  Grunwald, Cornelius; Bißmann, Stefan; Erdmann, Johannes; Hiller, Gudrun & Kröninger, Kevin
  
• Synergies of top and B anomalies in SMEFT, contribution to Moriond 2021
  S. Bißmann, …, K. Kröninger
  
• Top and beauty synergies in SMEFT-fits at present and future colliders. Journal of High Energy Physics, 2021(6).
  Bißmann, Stefan; Grunwald, Cornelius; Hiller, Gudrun & Kröninger, Kevin
  
• Reformulation of a likelihood approach to fake-lepton estimation in the framework of Bayesian inference. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 1021, 165939.
  Erdmann, Johannes; Grunwald, Cornelius; Kröninger, Kevin; La, Cagnina Salvatore; Röhrig, Lars & Varnes, Erich
  
• Synergies of Drell-Yan, top and beauty in global SMEFT fits, contribution to TOP 2022
  C. Grunwald, …, K. Kröninger, L. Nollen
  
• A Bayesian tune of the Herwig Monte Carlo event generator. Journal of Instrumentation, 18(10), P10033.
  La, Cagnina Salvatore; Kröninger, Kevin; Kluth, Stefan & Verbytskyi, Andrii