| | |
| | |
Stat |
Members: 3645 Articles: 2'500'096 Articles rated: 2609
18 April 2024 |
|
| | | |
|
Article overview
| |
|
Machine learning classification: case of Higgs boson CP state in H to tau tau$ decay at LHC | K. Lasocha
; E. Richter-Was
; D. Tracz
; Z. Was
; P. Winkowska
; | Date: |
19 Dec 2018 | Abstract: | The Machine Learning (ML) techniques are rapidly finding their place as
standard methods of the data analysis in High Energy Physics. In this paper we
continue discussion on their application to measurement of the CP state of the
Higgs boson discovered by Large Hadron Collider experiments at CERN laboratory
in 2012.
We consider measurement in the $H o au au$ decay channel and use ML
techniques to discriminate between models based on variables defined in the
multi-dimensional phase-space. We discuss and quantify possible improvements
for the two most sensitive decay modes: $ au^pm o
ho^pm
u$ with
$
ho^pm o pi^pm pi^0$ and $ au^pm o a_1^pm
u$ with $a_1^pm o
ho^0 pi^pm o 3 pi^pm$.
In previous publications information on the hadronic decay products of the
$ au$ leptons was used. Discriminating between Higgs boson CP state was
studied as binary classification problem. Now we show how approximate
constraints on the outgoing neutrinos momenta, not accessible in a direct way,
can help to improve classification performance. Added to the
ML clasification features significantly enhance the sensitivity for Higgs
boson CP state. In principle all information is provided with 4-momenta of the
final state particles present in the events. As we have observed in the past,
not all of such information is straightforward to be identified in ML training.
We investigate how optimised high-level features, like some angles of neutrino
orientation, may improve ML results. This can be understood as an intermediate
step toward choice of better classifiers where expert variables will not be
necessary.
For the performance comparison, in parallel to {it Deep Learning Neural
Network}, we use other ML methods: {it Boosted Trees}, {it Random Forest} and
{it Support Vector Machine}. | Source: | arXiv, 1812.8140 | Services: | Forum | Review | PDF | Favorites |
|
|
No review found.
Did you like this article?
Note: answers to reviews or questions about the article must be posted in the forum section.
Authors are not allowed to review their own article. They can use the forum section.
browser claudebot
|
| |
|
|
|
| News, job offers and information for researchers and scientists:
| |