IML Machine Learning Working Group

Tuesday 11 Oct 2022, 15:00 → 18:00 Europe/Zurich

40/S2-C01 - Salle Marie Sklodowska-Curie (CERN)

Topic: Optimal transport and invertible algorithms

15:00 → 15:05 News

Speakers: Anja Butter, Fabio Catalano (University and INFN Torino (IT)), Lorenzo Moneta (CERN), Michael Kagan (SLAC National Accelerator Laboratory (US)), Dr Pietro Vischia (Universite Catholique de Louvain (UCL) (BE)), Simon Akar (University of Cincinnati (US)), Stefano Carrazza (CERN)

IML_meeting_news_111022.pdf

15:05 → 15:30 Normalizing Flows for Differentiable Expectation Values

Speaker: Thorsten Glüsenkamp (Universität Erlangen-Nürnberg)

nfs_cern.pdf

15:30 → 15:35 Question time

15:35 → 16:00 Normalising Flows for Particle Cloud Generation

Speaker: Benno Kach (Deutsches Elektronen-Synchrotron (DE))

IML_Meeting16_9.pdf

16:00 → 16:05 Question time

16:05 → 16:30 Normalising Flows for Calorimeter Simulation

Speaker: Imahn Shekhzadeh (Haute école de gestion de Genève)

iml_workshop_imahn_shekhzadeh_11p10p22_pres.pdf

16:30 → 16:35 Question time

16:35 → 17:00 Two Invertible Networks for the Matrix Element Method

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.

Speaker: Theo Heimel (Heidelberg University)

INN_for_MEM_Theo_Heimel.pdf

17:00 → 17:05 Question time