-
Optimal Equivariant Architectures from the Symmetries of Matrix-Element Likelihoods
Authors:
Daniel Maître,
Vishal S. Ngairangbam,
Michael Spannowsky
Abstract:
The Matrix-Element Method (MEM) has long been a cornerstone of data analysis in high-energy physics. It leverages theoretical knowledge of parton-level processes and symmetries to evaluate the likelihood of observed events. In parallel, the advent of geometric deep learning has enabled neural network architectures that incorporate known symmetries directly into their design, leading to more effici…
▽ More
The Matrix-Element Method (MEM) has long been a cornerstone of data analysis in high-energy physics. It leverages theoretical knowledge of parton-level processes and symmetries to evaluate the likelihood of observed events. In parallel, the advent of geometric deep learning has enabled neural network architectures that incorporate known symmetries directly into their design, leading to more efficient learning. This paper presents a novel approach that combines MEM-inspired symmetry considerations with equivariant neural network design for particle physics analysis. Even though Lorentz invariance and permutation invariance overall reconstructed objects are the largest and most natural symmetry in the input domain, we find that they are sub-optimal in most practical search scenarios. We propose a longitudinal boost-equivariant message-passing neural network architecture that preserves relevant discrete symmetries. We present numerical studies demonstrating MEM-inspired architectures achieve new state-of-the-art performance in distinguishing di-Higgs decays to four bottom quarks from the QCD background, with enhanced sample and parameter efficiencies. This synergy between MEM and equivariant deep learning opens new directions for physics-informed architecture design, promising more powerful tools for probing physics beyond the Standard Model.
△ Less
Submitted 24 October, 2024;
originally announced October 2024.
-
Unweighting multijet event generation using factorisation-aware neural networks
Authors:
Timo Janßen,
Daniel Maître,
Steffen Schumann,
Frank Siegert,
Henry Truong
Abstract:
In this article we combine a recently proposed method for factorisation-aware matrix element surrogates with an unbiased unweighting algorithm. We show that employing a sophisticated neural network emulation of QCD multijet matrix elements based on dipole factorisation can lead to a drastic acceleration of unweighted event generation. We train neural networks for a selection of partonic channels c…
▽ More
In this article we combine a recently proposed method for factorisation-aware matrix element surrogates with an unbiased unweighting algorithm. We show that employing a sophisticated neural network emulation of QCD multijet matrix elements based on dipole factorisation can lead to a drastic acceleration of unweighted event generation. We train neural networks for a selection of partonic channels contributing at the tree-level to $Z+4,5$ jets and $t\bar{t}+3,4$ jets production at the LHC which necessitates a generalisation of the dipole emulation model to include initial state partons as well as massive final state quarks. We also present first steps towards the emulation of colour-sampled amplitudes. We incorporate these emulations as fast and accurate surrogates in a two-stage rejection sampling algorithm within the Sherpa Monte Carlo that yields unbiased unweighted events suitable for phenomenological analyses and post-processing in experimental workflows, e.g. as input to a time-consuming detector simulation. For the computational cost of unweighted events we achieve a reduction by factors between $16$ and $350$ for the considered channels.
△ Less
Submitted 16 May, 2023; v1 submitted 31 January, 2023;
originally announced January 2023.
-
Machine Learning and LHC Event Generation
Authors:
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
, et al. (26 additional authors not shown)
Abstract:
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 requi…
▽ More
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.
△ Less
Submitted 28 December, 2022; v1 submitted 14 March, 2022;
originally announced March 2022.
-
Les Houches 2019: Physics at TeV Colliders: Standard Model Working Group Report
Authors:
S. Amoroso,
P. Azzurri,
J. Bendavid,
E. Bothmann,
D. Britzger,
H. Brooks,
A. Buckley,
M. Calvetti,
X. Chen,
M. Chiesa,
L. Cieri,
V. Ciulli,
J. Cruz-Martinez,
A. Cueto,
A. Denner,
S. Dittmaier,
M. Donegà,
M. Dührssen-Debling,
I. Fabre,
S. Ferrario-Ravasio,
D. de Florian,
S. Forte,
P. Francavilla,
T. Gehrmann,
A. Gehrmann-De Ridder
, et al. (58 additional authors not shown)
Abstract:
This Report summarizes the proceedings of the 2019 Les Houches workshop on Physics at TeV Colliders. Session 1 dealt with (I) new developments for high precision Standard Model calculations, (II) the sensitivity of parton distribution functions to the experimental inputs, (III) new developments in jet substructure techniques and a detailed examination of gluon fragmentation at the LHC, (IV) issues…
▽ More
This Report summarizes the proceedings of the 2019 Les Houches workshop on Physics at TeV Colliders. Session 1 dealt with (I) new developments for high precision Standard Model calculations, (II) the sensitivity of parton distribution functions to the experimental inputs, (III) new developments in jet substructure techniques and a detailed examination of gluon fragmentation at the LHC, (IV) issues in the theoretical description of the production of Standard Model Higgs bosons and how to relate experimental measurements, and (V) Monte Carlo event generator studies relating to PDF evolution and comparisons of important processes at the LHC.
△ Less
Submitted 3 March, 2020;
originally announced March 2020.
-
Pushing the precision frontier at the LHC with V+jets
Authors:
Ulla Blumenschein,
Shane Breeze,
Chiara Debenedetti,
Engin Eren,
Simona Gargiulo,
Nigel Glover,
Jonas M. Lindert,
Daniel Maitre,
Sarah A. Malik,
Bjoern Penning,
Svenja K. Pflitsch,
Darren Price,
Stefan Richter,
Marek Schonherr,
Peter Schichter,
Paolo Torrielli,
Maria Ubiali,
Nicholas Wardle,
Angelo G. Zecchinelli
Abstract:
This documents the proceedings from a workshop titled `Illuminating Standard candles at the LHC: V+jets' held at Imperial College London on 25th-26th April 2017. It summarises the numerous contributions to the workshop, from the experimental overview of V+jets measurements at CMS and ATLAS and their role in searching for physics beyond the Standard Model to the status of higher order perturbative…
▽ More
This documents the proceedings from a workshop titled `Illuminating Standard candles at the LHC: V+jets' held at Imperial College London on 25th-26th April 2017. It summarises the numerous contributions to the workshop, from the experimental overview of V+jets measurements at CMS and ATLAS and their role in searching for physics beyond the Standard Model to the status of higher order perturbative calculations to these processes and their inclusion in state of the art Monte Carlo simulations. An executive summary of the ensuing discussions including a list of outcomes and wishlist for future consideration is also presented.
△ Less
Submitted 6 February, 2018;
originally announced February 2018.
-
Weak Vector Boson Production with Many Jets at the LHC $\sqrt{s}= 13$ TeV
Authors:
F. R. Anger,
F. Febres Cordero,
S. Höche,
D. Maître
Abstract:
Signatures with an electroweak vector boson and many jets play a crucial role at the Large Hadron Collider, both in the measurement of Standard-Model parameters and in searches for new physics. Precise predictions for these multi-scale processes are therefore indispensable. We present next-to-leading order QCD predictions for $W^\pm/Z$+jets at $\sqrt{s}=13$ TeV, including up to five/four jets in t…
▽ More
Signatures with an electroweak vector boson and many jets play a crucial role at the Large Hadron Collider, both in the measurement of Standard-Model parameters and in searches for new physics. Precise predictions for these multi-scale processes are therefore indispensable. We present next-to-leading order QCD predictions for $W^\pm/Z$+jets at $\sqrt{s}=13$ TeV, including up to five/four jets in the final state. All production channels are included and leptonic decays of the vector bosons are considered at the amplitude level. We assess theoretical uncertainties arising from renormalization- and factorization-scale dependence by considering fixed-order dynamical scales based on the $H_{\rm T}$ variable as well as on the MiNLO procedure. We also explore uncertainties associated to different choices of parton-distribution functions. We provide event samples that can be explored through publicly available $n$-tuple sets, generated with BlackHat in combination with SHERPA.
△ Less
Submitted 22 December, 2017;
originally announced December 2017.
-
Ntuples for NLO Events at Hadron Colliders
Authors:
Z. Bern,
L. J. Dixon,
F. Febres Cordero,
S. Hoeche,
H. Ita,
D. A. Kosower,
D. Maitre
Abstract:
We present an event-file format for the dissemination of next-to-leading-order (NLO) predictions for QCD processes at hadron colliders. The files contain all information required to compute generic jet-based infrared-safe observables at fixed order (without showering or hadronization), and to recompute observables with different factorization and renormalization scales. The files also make it poss…
▽ More
We present an event-file format for the dissemination of next-to-leading-order (NLO) predictions for QCD processes at hadron colliders. The files contain all information required to compute generic jet-based infrared-safe observables at fixed order (without showering or hadronization), and to recompute observables with different factorization and renormalization scales. The files also make it possible to evaluate cross sections and distributions with different parton distribution functions. This in turn makes it possible to estimate uncertainties in NLO predictions of a wide variety of observables without recomputing the short-distance matrix elements. The event files allow a user to choose among a wide range of commonly-used jet algorithms and jet-size parameters.
We provide event files for a $W$ or $Z$ boson accompanied by up to four jets, and for pure-jet events with up to four jets. The files are for the Large Hadron Collider with a center of mass energy of 7 or 8 TeV.
A C++ library along with a Python interface for handling these files are also provided and described in this article. The library allows a user to read the event files and recompute observables transparently for different pdf sets and factorization and renormalization scales.
△ Less
Submitted 28 October, 2013;
originally announced October 2013.
-
The SM and NLO Multileg and SM MC Working Groups: Summary Report
Authors:
J. Alcaraz Maestre,
S. Alioli,
J. R. Andersen,
R. D. Ball,
A. Buckley,
M. Cacciari,
F. Campanario,
N. Chanon,
G. Chachamis,
V. Ciulli,
F. Cossutti,
G. Cullen,
A. Denner,
S. Dittmaier,
J. Fleischer,
R. Frederix,
S. Frixione,
J. Gao,
L. Garren,
S. Gascon-Shotkin,
N. Greiner,
J. P. Guillet,
T. Hapola,
N. P. Hartland,
G. Heinrich
, et al. (57 additional authors not shown)
Abstract:
The 2011 Les Houches workshop was the first to confront LHC data. In the two years since the previous workshop there have been significant advances in both soft and hard QCD, particularly in the areas of multi-leg NLO calculations, the inclusion of those NLO calculations into parton shower Monte Carlos, and the tuning of the non-perturbative parameters of those Monte Carlos. These proceedings desc…
▽ More
The 2011 Les Houches workshop was the first to confront LHC data. In the two years since the previous workshop there have been significant advances in both soft and hard QCD, particularly in the areas of multi-leg NLO calculations, the inclusion of those NLO calculations into parton shower Monte Carlos, and the tuning of the non-perturbative parameters of those Monte Carlos. These proceedings describe the theoretical advances that have taken place, the impact of the early LHC data, and the areas for future development.
△ Less
Submitted 9 May, 2012; v1 submitted 29 March, 2012;
originally announced March 2012.
-
Four-Jet Production at the Large Hadron Collider at Next-to-Leading Order in QCD
Authors:
Z. Bern,
G. Diana,
L. J. Dixon,
F. Febres Cordero,
S. Hoeche,
D. A. Kosower,
H. Ita,
D. Maitre,
K. Ozeren
Abstract:
We present the cross sections for production of up to four jets at the Large Hadron Collider, at next-to-leading order in the QCD coupling. We use the BlackHat library in conjunction with SHERPA and a recently developed algorithm for assembling primitive amplitudes into color-dressed amplitudes. We adopt the cuts used by ATLAS in their study of multi-jet events in pp collisions at \sqrt{s} = 7 TeV…
▽ More
We present the cross sections for production of up to four jets at the Large Hadron Collider, at next-to-leading order in the QCD coupling. We use the BlackHat library in conjunction with SHERPA and a recently developed algorithm for assembling primitive amplitudes into color-dressed amplitudes. We adopt the cuts used by ATLAS in their study of multi-jet events in pp collisions at \sqrt{s} = 7 TeV. We include estimates of nonperturbative corrections and compare to ATLAS data. We store intermediate results in a framework that allows the inexpensive computation of additional results for different choices of scale or parton distributions.
△ Less
Submitted 21 July, 2012; v1 submitted 16 December, 2011;
originally announced December 2011.
-
Precise Predictions for W + 4 Jet Production at the Large Hadron Collider
Authors:
C. F. Berger,
Z. Bern,
L. J. Dixon,
F. Febres Cordero,
D. Forde,
T. Gleisberg,
H. Ita,
D. A. Kosower,
D. Maitre
Abstract:
We present the first next-to-leading order QCD results for W + 4-jet production at hadron colliders. Total cross sections, as well as distributions in the jet transverse momenta and in the total transverse energy H_T are provided for the initial LHC energy of \sqrt{s} = 7 TeV. We use a leading-color approximation, known to be accurate to 3% for W production with fewer jets. The virtual matrix elem…
▽ More
We present the first next-to-leading order QCD results for W + 4-jet production at hadron colliders. Total cross sections, as well as distributions in the jet transverse momenta and in the total transverse energy H_T are provided for the initial LHC energy of \sqrt{s} = 7 TeV. We use a leading-color approximation, known to be accurate to 3% for W production with fewer jets. The virtual matrix elements and the most complicated real-emission matrix elements are handled by the BlackHat library, based on on-shell methods. The remaining parts of the calculation, including the integration over phase space, are performed by the SHERPA package.
△ Less
Submitted 10 February, 2011; v1 submitted 13 September, 2010;
originally announced September 2010.
-
Vector Boson + Jets with BlackHat and Sherpa
Authors:
C. F. Berger,
Z. Bern,
L. J. Dixon,
F. Febres Cordero,
D. Forde,
T. Gleisberg,
H. Ita,
D. A. Kosower,
D. Maitre
Abstract:
We review recent NLO QCD results for W,Z + 3-jet production at hadron colliders, computed using BlackHat and SHERPA. We also include some new results for Z + 3-jet production at the LHC at 7 TeV. We report new progress towards the NLO cross section for W + 4-jet production. In particular, we show that the virtual matrix elements produced by BlackHat are numerically stable. We also show that with a…
▽ More
We review recent NLO QCD results for W,Z + 3-jet production at hadron colliders, computed using BlackHat and SHERPA. We also include some new results for Z + 3-jet production at the LHC at 7 TeV. We report new progress towards the NLO cross section for W + 4-jet production. In particular, we show that the virtual matrix elements produced by BlackHat are numerically stable. We also show that with an improved integrator and tree-level matrix elements from BlackHat, SHERPA produces well-behaved real-emission contributions. As an illustration, we present the real-emission contributions -- including dipole-subtraction terms -- to the p_T distribution of the fourth jet, for a single subprocess with the maximum number of gluons.
△ Less
Submitted 20 May, 2010;
originally announced May 2010.
