Application of a Convolutional Neural Network for image classification for the analysis of collisions in High Energy Physics

Abstract

The application of deep learning techniques using convolutional neural networks for the classification of particle collisions in High Energy Physics is explored. An intuitive approach to transform physical variables, like momenta of particles and jets, into a single image that captures the relevant information, is proposed. The idea is tested using a well-known deep learning framework on a simulation dataset, including leptonic ttbar events and the corresponding background at 7 TeV from the CMS experiment at LHC, available as Open Data. This initial test shows competitive results when compared to more classical approaches, like those using feedforward neural networks.

Searching in CMS open data for dimuon resonances with substantial transverse momentum

“…[58] the high-p T tail of the data below m µµ < 50 GeV. 6 The CMS11a data set and the DYMC/ZMC samples show fairly good agreement in Fig. 1, including the shape of the Z pole and the shape of the trigger turn-on region.…”

“…The latter three uncertainties, which are essentially uniform across our mass range, are combined together in quadrature and assessed, after the resolution profiling, using the multiplicative approach in Eq. (6). (Because the uncertainty in V IP is so subdominant, its presence in the prompt samples does not alter the incremental effect on the expected limits.)…”

“…Where the DYMC and ZMC samples meet at m µµ = 50 GeV, there is a small mismatch, again within the expected theoretical uncertainties of the simulations. (Strictly speaking, the DYMC and ZMC samples are defined by the 6 We must rely on the unmatched Pythia-only DY sample here, because the 2011 CMS Open Data release provided Mad-Graph/Pythia matched samples for DY plus {1, 2, 3, 4} jets [64][65][66][67] but not for DY plus 0 jets. This highlights the importance of stress-testing archival data strategies, to ensure that relevant information is not inadvertently omitted.…”

“…[4,5]; see Refs. [6][7][8] for machine-learning studies on detector-simulated CMS samples, Refs. [9][10][11] for QCD studies on archival ALEPH data, and Ref.…”

“…, the line-shape uncertainty dominates, leading to a shift in the expected limits of around 0.6%. Note that Eq (6). …”