Speaker
Dr
Jörg Stelzer
(CERN)
Description
In high-energy physics, with the search for ever smaller signals in
ever larger data sets, it has become essential to extract a maximum of
the available information from the data. Multivariate classification
methods based on machine learning techniques have become a fundamental
ingredient to most analyses. Also the multivariate classifiers
themselves have significantly evolved in recent years. Statisticians
have found new ways to tune and to combine classifiers to further gain
in performance. Integrated into the analysis framework ROOT, TMVA is
a toolkit which holds a large variety of multivariate classification
algorithms. They range from rectangular cut optimization using a
genetic algorithm and from likelihood estimators over the linear
Fisher discriminant and non-linear neural networks, to sophisticated
methods like support vector machines, boosted decision trees and rule
ensemble fitting that was recently developed. TMVA manages the
simultaneous training, testing, and performance evaluation of all these
classifiers with a user-friendly interface, and expedites the application
of the trained classifiers to data.
Primary authors
Dr
Andreas Hoecker
(CERN, Switzerland)
Dr
Fedrik Tegenfeldt
(Iowa University, USA)
Dr
Helge Voss
(MPI for Nuclear Physics, Heidelberg, Germany)
Dr
Jörg Stelzer
(CERN)
