IML Machine Learning Working Group
Tuesday 11 October 2022 -
15:00
Monday 10 October 2022
Tuesday 11 October 2022
15:00
News
-
Stefano Carrazza
(
CERN
)
Fabio Catalano
(
University and INFN Torino (IT)
)
Anja Butter
Lorenzo Moneta
(
CERN
)
Pietro Vischia
(
Universite Catholique de Louvain (UCL) (BE)
)
Simon Akar
(
University of Cincinnati (US)
)
Michael Kagan
(
SLAC National Accelerator Laboratory (US)
)
News
Stefano Carrazza
(
CERN
)
Fabio Catalano
(
University and INFN Torino (IT)
)
Anja Butter
Lorenzo Moneta
(
CERN
)
Pietro Vischia
(
Universite Catholique de Louvain (UCL) (BE)
)
Simon Akar
(
University of Cincinnati (US)
)
Michael Kagan
(
SLAC National Accelerator Laboratory (US)
)
15:00 - 15:05
Room: 40/S2-C01 - Salle Curie
15:05
Normalizing Flows for Differentiable Expectation Values
-
Thorsten Glüsenkamp
(
Universität Erlangen-Nürnberg
)
Normalizing Flows for Differentiable Expectation Values
Thorsten Glüsenkamp
(
Universität Erlangen-Nürnberg
)
15:05 - 15:30
Room: 40/S2-C01 - Salle Curie
15:30
Question time
Question time
15:30 - 15:35
Room: 40/S2-C01 - Salle Curie
15:35
Normalising Flows for Particle Cloud Generation
-
Benno Kach
(
Deutsches Elektronen-Synchrotron (DE)
)
Normalising Flows for Particle Cloud Generation
Benno Kach
(
Deutsches Elektronen-Synchrotron (DE)
)
15:35 - 16:00
Room: 40/S2-C01 - Salle Curie
16:00
Question time
Question time
16:00 - 16:05
Room: 40/S2-C01 - Salle Curie
16:05
Normalising Flows for Calorimeter Simulation
-
Imahn Shekhzadeh
(
Haute école de gestion de Genève
)
Normalising Flows for Calorimeter Simulation
Imahn Shekhzadeh
(
Haute école de gestion de Genève
)
16:05 - 16:30
Room: 40/S2-C01 - Salle Curie
16:30
Question time
Question time
16:30 - 16:35
Room: 40/S2-C01 - Salle Curie
16:35
Two Invertible Networks for the Matrix Element Method
-
Theo Heimel
(
Heidelberg University
)
Two Invertible Networks for the Matrix Element Method
Theo Heimel
(
Heidelberg University
)
16:35 - 17:00
Room: 40/S2-C01 - Salle Curie
The matrix element method is widely considered the ultimate LHC inference tool for small event numbers, but computationally expensive. We show how a combination of two conditional generative neural networks encodes the QCD radiation and detector effects without any simplifying assumptions and allows us to efficiently compute the likelihood for individual hard-scattering events. We illustrate our approach for the CP-violating phase of the top Yukawa coupling in associated Higgs and single-top production. The limiting factor for the precision of our approach is jet combinatorics.
17:00
Question time
Question time
17:00 - 17:05
Room: 40/S2-C01 - Salle Curie