10-15 March 2019
Steinmatte conference center
Europe/Zurich timezone

Incorporation of Systematic Uncertainties in the Training of Multivariate Methods

13 Mar 2019, 15:30
Steinmatte Room A

Steinmatte Room A

Oral Track 2: Data Analysis - Algorithms and Tools Track 2: Data Analysis - Algorithms and Tools


Thomas Alef (University of Bonn (DE))


Multivariate analyses in particle physics often reach a precision such that its uncertainties are dominated by systematic effects. While there are known strategies to mitigate systematic effects based on adversarial neural nets, the application of Boosted Decision Trees (BDT) so far had to ignore systematics in the training.
We present a method to incorporate systematic uncertainties into a BDT, the "systematics-aware BDT" (saBDT).
We evaluate our method on open data of the ATLAS Higgs to tau tau machine learning challenge and compare our results to neural nets trained with an adversary to mitigate systematic effects.

Primary author

Thomas Alef (University of Bonn (DE))


Eckhard Von Torne (University of Bonn (DE))

Presentation materials

Peer reviewing