Data Analysis in HEP

Data Analysis in Experimental and High-Energy Physics

Members of our research group have been actively involved in the analysis of data collected by former DØ experiment at the Fermi National Accelerator Laboratory near Chicago, USA. We participated at several analysis and beside using TMVA and other analytic tools common in HEP, we started to develop new machine learning methods for signal identification and tests for MC and data comparison. Since 2016 we are part of the extensive research infrastructure of the Czech Ministry of Education, Youth and Sports for cooperation with experiments in Fermilab. Nowadays our team works closely with neutrino experimetns NOvA and DUNE. As part of these projects, we have already sent dozens students to the US where they worked at the Fermilab  laboratories. In addition to cooperation with FNAL we cooperate intensively with ATLAS experiment in CERN.

Within the GAMS group, we focus primarily on developing new statistical methods for separating signal from background.  Since the separation problem in high energy physics is limited to a few standard methods, such as decision tree or neural networks, our goal is to grasp other approaches from other areas and apply them to physical data. Therefore, we are studying and using separation techniques such as cluster modeling (MBC), generalized linear models (GLM) or Phi-divergence supervised decision trees (Phi-SDDT). Further we are working on Monte Carlo (MC) quality testing, where we are creating new distribution tests, which can be included in ROOT framework. The last version of genarilezed homogeneity test for weighted data samples ca nbe downloaded here.

Our department takes part in research infrastructure FERMILAB-CZ.


  • Experiments DØ, NOvA, and DUNE, Fermi National Accelerator Laboratory (Fermilab), Batavia, Illinois, USA.
  • Experiment ATLAS, CERN, Geneve, Switzerland.
  • Department of Particle Theory and Phenomenology, Institute of Physics, Academy of Sciences of the Czech Republic.
  • Department of Optimization and Systems, Institute of Computer Science, Academy of Sciences of the Czech Republic.


Jiří Franc

assistant professor at FNSPE CTU in Prague

Václav Kůs

assistant professor at FNSPE CTU in Prague

Petr Vokáč

research and technical associate at FNSPE CTU in Prague

Zdeněk Čulík

assistant professor at FNSPE CTU in Prague

Petr Bouř

Ph.D. student at FNSPE CTU in Prague

(divergent statistical methods and their applications in elementary particle physics)

Kristina Jarůšková

Ph.D. student at FNSPE CTU in Prague

(statistical separation and identification by means of divergence techniques for multi-dimensional data)



Michal Štěpánek

(application of machine learning algorithms in highenergy physics)

Petr Jačka

(measurement of kinematic properties of top-antitop pairs)

Jakub Trusina

(study of homogeneity tests’ properties on weighted data samples)

Adam Novotný

(Application of statistical tests of homogeneity on data from HEP and DUNE neutrino flavour classification using a CNN )

Miroslav Kubů

(Convolutional Neural Networks in High Energy Physics and NOvA Experiment)

Kateřina Hladká

(Machine signal separation from decays of heavy d+Au nuclei in the STAR/BNL Experiment)





Significant papers

  • D0 Collaboration (Abazov, Victor Mukhamedovich et al.)  Measurement of the inclusive ttbar production cross section in ppbar collisions at ?s=1.96 TeV and determination of the top quark pole mass – Phys.Rev. D94 (2016) 092004.
  • Štěpánek, M. – Franc, J. – Kůs, V.: Model cluster method as a new multivariable technique in high energy physics. Journal of Physics: Conference Series. 2014, ISSN 1742-6588.
  • Franc, J. – Bouř, P. – Štěpánek, M. – Kůs, V .: New Statistical Techniques in Measurement of Cross-Section Production by Ink of the Best Pairs. Within the Top 2014 Conference. Stanford: Stanford University, 2014.
  • Štěpánek, M. – Franc, J. – Kůs, V .: Modification of Gauss mixtures for classification of data in high energy physics. Journal of Physics: Conference Series. 2015, sv. 574, No. 1, Art. 012

Bachelor and Diploma projects for students