Contributor info: Dr Anja Butter

Theo Heimel, Nathan Huetsch, Ramon Winterhalder, Tilman Plehn, Anja Butter

SciPost Phys. 17, 129 (2024) · published 8 November 2024 | Toggle abstract · pdf

The matrix element method is the LHC inference method of choice for limited statistics. We present a dedicated machine learning framework, based on efficient phase-space integration, a learned acceptance and transfer function. It is based on a choice of INN and diffusion networks, and a transformer to solve jet combinatorics. We showcase this setup for the CP-phase of the top Yukawa coupling in associated Higgs and single-top production.

Anja Butter, Tomáš Ježo, Michael Klasen, Mathias Kuschick, Sofia Palacios Schweitzer, Tilman Plehn

SciPost Phys. Core 7, 064 (2024) · published 16 September 2024 | Toggle abstract · pdf

Off-shell effects in large LHC backgrounds are crucial for precision predictions and, at the same time, challenging to simulate. We present a novel method to transform high-dimensional distributions based on a diffusion neural network and use it to generate a process with off-shell kinematics from the much simpler on-shell one. Applied to a toy example of top pair production at LO we show how our method generates off-shell configurations fast and precisely, while reproducing even challenging on-shell features.

Mathias Backes, Anja Butter, Monica Dunford, Bogdan Malaescu

SciPost Phys. Core 7, 007 (2024) · published 21 February 2024 | Toggle abstract · pdf

The unfolding of detector effects is crucial for the comparison of data to theory predictions. While traditional methods are limited to representing the data in a low number of dimensions, machine learning has enabled new unfolding techniques while retaining the full dimensionality. Generative networks like invertible neural networks~(INN) enable a probabilistic unfolding, which map individual data events to their corresponding unfolded probability distribution. The accuracy of such methods is however limited by how well simulated training samples model the actual data that is unfolded. We introduce the iterative conditional INN (IcINN) for unfolding that adjusts for deviations between simulated training samples and data. The IcINN unfolding is first validated on toy data and then applied to pseudo-data for the $pp \to Z \gamma \gamma$ process.

Anja Butter, Tilman Plehn, Nathalie Soybelman, Johann Brehmer

SciPost Phys. 16, 037 (2024) · published 29 January 2024 | Toggle abstract · pdf

While neural networks offer an attractive way to numerically encode functions, actual formulas remain the language of theoretical particle physics. We use symbolic regression trained on matrix-element information to extract, for instance, optimal LHC observables. This way we invert the usual simulation paradigm and extract easily interpretable formulas from complex simulated data. We introduce the method using the effect of a dimension-6 coefficient on associated ZH production. We then validate it for the known case of CP-violation in weak-boson-fusion Higgs production, including detector effects.

Anja Butter, Barry M. Dillon, Tilman Plehn, Lorenz Vogel

SciPost Phys. Core 6, 085 (2023) · published 12 December 2023 | Toggle abstract · pdf

Discriminating quark-like from gluon-like jets is, in many ways, a key challenge for many LHC analyses. First, we use a known difference in PYTHIA and HERWIG simulations to show how decorrelated taggers would break down when the most distinctive feature is aligned with theory uncertainties. We propose conditional training on interpolated samples, combined with a controlled Bayesian network, as a more resilient framework. The interpolation parameter can be used to optimize the training evaluated on a calibration dataset, and to test the stability of this optimization. The interpolated training might also be useful to track generalization errors when training networks on simulations.

Theo Heimel, Ramon Winterhalder, Anja Butter, Joshua Isaacson, Claudius Krause, Fabio Maltoni, Olivier Mattelaer, Tilman Plehn

SciPost Phys. 15, 141 (2023) · published 6 October 2023 | Toggle abstract · pdf

Theory predictions for the LHC require precise numerical phase-space integration and generation of unweighted events. We combine machine-learned multi-channel weights with a normalizing flow for importance sampling, to improve classical methods for numerical integration. We develop an efficient bi-directional setup based on an invertible network, combining online and buffered training for potentially expensive integrands. We illustrate our method for the Drell-Yan process with an additional narrow resonance.

Anja Butter, Theo Heimel, Till Martini, Sascha Peitzsch, Tilman Plehn

SciPost Phys. 15, 094 (2023) · published 14 September 2023 | Toggle abstract · pdf

The matrix element method is widely considered the ultimate LHC inference tool for small event numbers. We show how a combination of two conditional generative neural networks encodes the QCD radiation and detector effects without any simplifying assumptions, while keeping the computation of likelihoods for individual events numerically efficient. We illustrate our approach for the CP-violating phase of the top Yukawa coupling in associated Higgs and single-top production. Currently, the limiting factor for the precision of our approach is jet combinatorics.

Simon Badger, Anja Butter, Michel Luchmann, Sebastian Pitz, Tilman Plehn

SciPost Phys. Core 6, 034 (2023) · published 25 April 2023 | Toggle abstract · pdf

Evaluating loop amplitudes is a time-consuming part of LHC event generation. For di-photon production with jets we show that simple, Bayesian networks can learn such amplitudes and model their uncertainties reliably. A boosted training of the Bayesian network further improves the uncertainty estimate and the network precision in critical phase space regions. In general, boosted network training of Bayesian networks allows us to move between fit-like and interpolation-like regimes of network training.

Anja Butter, Tilman Plehn, Steffen Schumann, Simon Badger, Sascha Caron, Kyle Cranmer, Francesco Armando Di Bello, Etienne Dreyer, Stefano Forte, Sanmay Ganguly, Dorival Gonçalves, Eilam Gross, Theo Heimel, Gudrun Heinrich, Lukas Heinrich, Alexander Held, Stefan Höche, Jessica N. Howard, Philip Ilten, Joshua Isaacson, Timo Janßen, Stephen Jones, Marumi Kado, Michael Kagan, Gregor Kasieczka, Felix Kling, Sabine Kraml, Claudius Krause, Frank Krauss, Kevin Kröninger, Rahool Kumar Barman, Michel Luchmann, Vitaly Magerya, Daniel Maitre, Bogdan Malaescu, Fabio Maltoni, Till Martini, Olivier Mattelaer, Benjamin Nachman, Sebastian Pitz, Juan Rojo, Matthew Schwartz, David Shih, Frank Siegert, Roy Stegeman, Bob Stienen, Jesse Thaler, Rob Verheyen, Daniel Whiteson, Ramon Winterhalder, Jure Zupan

SciPost Phys. 14, 079 (2023) · published 21 April 2023 | Toggle abstract · pdf

First-principle simulations are at the heart of the high-energy physics research program. They link the vast data output of multi-purpose detectors with fundamental theory predictions and interpretation. This review illustrates a wide range of applications of modern machine learning to event generation and simulation-based inference, including conceptional developments driven by the specific requirements of particle physics. New ideas and tools developed at the interface of particle physics and machine learning will improve the speed and precision of forward simulations, handle the complexity of collision data, and enhance inference as an inverse simulation problem.

Anja Butter, Theo Heimel, Sander Hummerich, Tobias Krebs, Tilman Plehn, Armand Rousselot, Sophia Vent

SciPost Phys. 14, 078 (2023) · published 20 April 2023 | Toggle abstract · pdf

Generative networks are opening new avenues in fast event generation for the LHC. We show how generative flow networks can reach percent-level precision for kinematic distributions, how they can be trained jointly with a discriminator, and how this discriminator improves the generation. Our joint training relies on a novel coupling of the two networks which does not require a Nash equilibrium. We then estimate the generation uncertainties through a Bayesian network setup and through conditional data augmentation, while the discriminator ensures that there are no systematic inconsistencies compared to the training data.

Anja Butter, Sascha Diefenbacher, Gregor Kasieczka, Benjamin Nachman, Tilman Plehn, David Shih, Ramon Winterhalder

SciPost Phys. 13, 087 (2022) · published 7 October 2022 | Toggle abstract · pdf

The large data rates at the LHC require an online trigger system to select relevant collisions. Rather than compressing individual events, we propose to compress an entire data set at once. We use a normalizing flow as a deep generative model to learn the probability density of the data online. The events are then represented by the generative neural network and can be inspected offline for anomalies or used for other analysis purposes. We demonstrate our new approach for a toy model and a correlation-enhanced bump hunt.

Pierre Baldi, Lukas Blecher, Anja Butter, Julian Collado, Jessica N. Howard, Fabian Keilbach, Tilman Plehn, Gregor Kasieczka, Daniel Whiteson

SciPost Phys. 13, 064 (2022) · published 23 September 2022 | Toggle abstract · pdf

QCD-jets at the LHC are described by simple physics principles. We show how super-resolution generative networks can learn the underlying structures and use them to improve the resolution of jet images. We test this approach on massless QCD-jets and on fat top-jets and find that the network reproduces their main features even without training on pure samples. In addition, we show how a slim network architecture can be constructed once we have control of the full network performance.

Ramon Winterhalder, Vitaly Magerya, Emilio Villa, Stephen P. Jones, Matthias Kerner, Anja Butter, Gudrun Heinrich, Tilman Plehn

SciPost Phys. 12, 129 (2022) · published 13 April 2022 | Toggle abstract · pdf

Numerical evaluations of Feynman integrals often proceed via a deformation of the integration contour into the complex plane. While valid contours are easy to construct, the numerical precision for a multi-loop integral can depend critically on the chosen contour. We present methods to optimize this contour using a combination of optimized, global complex shifts and a normalizing flow. They can lead to a significant gain in precision.

Anja Butter, Sascha Diefenbacher, Gregor Kasieczka, Benjamin Nachman, Tilman Plehn

SciPost Phys. 10, 139 (2021) · published 10 June 2021 | Toggle abstract · pdf

A critical question concerning generative networks applied to event generation in particle physics is if the generated events add statistical precision beyond the training sample. We show for a simple example with increasing dimensionality how generative networks indeed amplify the training statistics. We quantify their impact through an amplification factor or equivalent numbers of sampled events.

Sebastian Bieringer, Anja Butter, Theo Heimel, Stefan Höche, Ullrich Köthe, Tilman Plehn, Stefan T. Radev

SciPost Phys. 10, 126 (2021) · published 2 June 2021 | Toggle abstract · pdf

QCD splittings are among the most fundamental theory concepts at the LHC. We show how they can be studied systematically with the help of invertible neural networks. These networks work with sub-jet information to extract fundamental parameters from jet samples. Our approach expands the LEP measurements of QCD Casimirs to a systematic test of QCD properties based on low-level jet observables. Starting with an toy example we study the effect of the full shower, hadronization, and detector effects in detail.

Mathias Backes, Anja Butter, Tilman Plehn, Ramon Winterhalder

SciPost Phys. 10, 089 (2021) · published 23 April 2021 | Toggle abstract · pdf

Event generation with neural networks has seen significant progress recently. The big open question is still how such new methods will accelerate LHC simulations to the level required by upcoming LHC runs. We target a known bottleneck of standard simulations and show how their unweighting procedure can be improved by generative networks. This can, potentially, lead to a very significant gain in simulation speed.

Marco Bellagente, Anja Butter, Gregor Kasieczka, Tilman Plehn, Armand Rousselot, Ramon Winterhalder, Lynton Ardizzone, Ullrich Köthe

SciPost Phys. 9, 074 (2020) · published 18 November 2020 | Toggle abstract · pdf

For simulations where the forward and the inverse directions have a physics meaning, invertible neural networks are especially useful. A conditional INN can invert a detector simulation in terms of high-level observables, specifically for ZW production at the LHC. It allows for a per-event statistical interpretation. Next, we allow for a variable number of QCD jets. We unfold detector effects and QCD radiation to a pre-defined hard process, again with a per-event probabilistic interpretation over parton-level phase space.

Anja Butter, Tilman Plehn, Ramon Winterhalder

SciPost Phys. Core 3, 009 (2020) · published 5 November 2020 | Toggle abstract · pdf

Subtracting event samples is a common task in LHC simulation and analysis, and standard solutions tend to be inefficient. We employ generative adversarial networks to produce new event samples with a phase space distribution corresponding to added or subtracted input samples. We first illustrate for a toy example how such a network beats the statistical limitations of the training data. We then show how such a network can be used to subtract background events or to include non-local collinear subtraction events at the level of unweighted 4-vector events.

Marco Bellagente, Anja Butter, Gregor Kasieczka, Tilman Plehn, Ramon Winterhalder

SciPost Phys. 8, 070 (2020) · published 29 April 2020 | Toggle abstract · pdf

LHC analyses directly comparing data and simulated events bear the danger of using first-principle predictions only as a black-box part of event simulation. We show how simulations, for instance, of detector effects can instead be inverted using generative networks. This allows us to reconstruct parton level information from measured events. Our results illustrate how, in general, fully conditional generative networks can statistically invert Monte Carlo simulations. As a technical by-product we show how a maximum mean discrepancy loss can be staggered or cooled.

Anja Butter, Tilman Plehn, Ramon Winterhalder

SciPost Phys. 7, 075 (2019) · published 4 December 2019 | Toggle abstract · pdf

Event generation for the LHC can be supplemented by generative adversarial networks, which generate physical events and avoid highly inefficient event unweighting. For top pair production we show how such a network describes intermediate on-shell particles, phase space boundaries, and tails of distributions. In particular, we introduce the maximum mean discrepancy to resolve sharp local features. It can be extended in a straightforward manner to include for instance off-shell contributions, higher orders, or approximate detector effects.

Gregor Kasieczka, Tilman Plehn, Anja Butter, Kyle Cranmer, Dipsikha Debnath, Barry M. Dillon, Malcolm Fairbairn, Darius A. Faroughy, Wojtek Fedorko, Christophe Gay, Loukas Gouskos, Jernej F. Kamenik, Patrick T. Komiske, Simon Leiss, Alison Lister, Sebastian Macaluso, Eric M. Metodiev, Liam Moore, Ben Nachman, Karl Nordström, Jannicke Pearkes, Huilin Qu, Yannik Rath, Marcel Rieger, David Shih, Jennifer M. Thompson, Sreedevi Varma

SciPost Phys. 7, 014 (2019) · published 30 July 2019 | Toggle abstract · pdf

Based on the established task of identifying boosted, hadronically decaying top quarks, we compare a wide range of modern machine learning approaches. Unlike most established methods they rely on low-level input, for instance calorimeter output. While their network architectures are vastly different, their performance is comparatively similar. In general, we find that these new approaches are extremely powerful and great fun.

Elias Bernreuther, Jan Horak, Tilman Plehn, Anja Butter

SciPost Phys. 5, 034 (2018) · published 17 October 2018 | Toggle abstract · pdf

Mono-X searches are standard dark matter search strategies at the LHC. First, we show how in the case of initial state radiation they essentially collapse to mono-jet searches. Second, we systematically study mono-X signatures from decays of heavier dark matter states. Direct detection constraints strongly limit our MSSM expectations, but largely vanish for mono-Z and mono-Higgs signals once we include light NMSSM mediators. Finally, the decay topology motivates mono-W-pair and mono-Higgs-pair searches, strengthening and complementing their mono-X counterparts.

Anja Butter, Gregor Kasieczka, Tilman Plehn, Michael Russell

SciPost Phys. 5, 028 (2018) · published 26 September 2018 | Toggle abstract · pdf

We introduce a new and highly efficient tagger for hadronically decaying top quarks, based on a deep neural network working with Lorentz vectors and the Minkowski metric. With its novel machine learning setup and architecture it allows us to identify boosted top quarks not only from calorimeter towers, but also including tracking information. We show how the performance of our tagger compares with QCD-inspired and image-recognition approaches and find that it significantly increases the performance for strongly boosted top quarks.