Contributor info: Mr Michel Luchmann

Marco Bellagente, Manuel Haußmann, Michel Luchmann, Tilman Plehn

SciPost Phys. 13, 003 (2022) · published 21 July 2022 | Toggle abstract · pdf

Following the growing success of generative neural networks in LHC simulations, the crucial question is how to control the networks and assign uncertainties to their event output. We show how Bayesian normalizing flow or invertible networks capture uncertainties from the training and turn them into an uncertainty on the event weight. Fundamentally, the interplay between density and uncertainty estimates indicates that these networks learn functions in analogy to parameter fits rather than binned event counts.

Ilaria Brivio, Sebastian Bruggisser, Emma Geoffray, Wolfgang Kilian, Michael Krämer, Michel Luchmann, Tilman Plehn, Benjamin Summ

SciPost Phys. 12, 036 (2022) · published 24 January 2022 | Toggle abstract · pdf

We present a global analysis of the Higgs and electroweak sector, in the SMEFT framework and matched to a UV-completion. As the UV-model we use the triplet extension of the electroweak gauge sector. The matching is performed at one loop, employing functional methods. In the SFitter analysis, we pay particular attention to theory uncertainties arising from the matching. Our results highlight the complementarity between SMEFT and model-specific analyses.

Gregor Kasieczka, Michel Luchmann, Florian Otterpohl, Tilman Plehn

SciPost Phys. 9, 089 (2020) · published 18 December 2020 | Toggle abstract · pdf

We show how to treat systematic uncertainties using Bayesian deep networks for regression. First, we analyze how these networks separately trace statistical and systematic uncertainties on the momenta of boosted top quarks forming fat jets. Next, we propose a novel calibration procedure by training on labels and their error bars. Again, the network cleanly separates the different uncertainties. As a technical side effect, we show how Bayesian networks can be extended to describe non-Gaussian features.

Sven Bollweg, Manuel Haussmann, Gregor Kasieczka, Michel Luchmann, Tilman Plehn, Jennifer Thompson

SciPost Phys. 8, 006 (2020) · published 16 January 2020 | Toggle abstract · pdf

Bayesian neural networks allow us to keep track of uncertainties, for example in top tagging, by learning a tagger output together with an error band. We illustrate the main features of Bayesian versions of established deep-learning taggers. We show how they capture statistical uncertainties from finite training samples, systematics related to the jet energy scale, and stability issues through pile-up. Altogether, Bayesian networks offer many new handles to understand and control deep learning at the LHC without introducing a visible prior effect and without compromising the network performance.