44 research outputs found
Josef Winterhalder jr. - Illusive altars, Dačice-Jemnice-Běhařovice-Trstěnice
No full textThis bachelor thesis deals with the work of painter Josef Winterhalder Jr. with a focus on creation of illusive altars. Places with these altars were selected with regard to two criteria, namely, the church is not convent and the fresco was not created in collaboration with another painter. These churches are sv. Vavřince in Dačice, sv. Víta in Jemnice, Nejsvětější trojice in Běhařovice and Povýšení sv. Kříže in Třeštice. Based on the iconographic and formal analysis, the fresco altar work of the author will be evaluated in the historical context of the location. Keywords Josef Winterhalder ml., baroque, illusive altar, fresco, Dačice, Jemnice, Běhařovice, Trstěnic
BitHEP -- The Limits of Low-Precision ML in HEP
No full textThe increasing complexity of modern neural network architectures demands fast
and memory-efficient implementations to mitigate computational bottlenecks. In
this work, we evaluate the recently proposed BitNet architecture in HEP
applications, assessing its performance in classification, regression, and
generative modeling tasks. Specifically, we investigate its suitability for
quark-gluon discrimination, SMEFT parameter estimation, and detector
simulation, comparing its efficiency and accuracy to state-of-the-art methods.
Our results show that while BitNet consistently performs competitively in
classification tasks, its performance in regression and generation varies with
the size and type of the network, highlighting key limitations and potential
areas for improvement
The MadNIS Reloaded
No full textIn pursuit of precise and fast theory predictions for the LHC, we present an implementation of the MadNIS method in the MadGraph event generator. A series of improvements in MadNIS further enhance its efficiency and speed. We validate this implementation for realistic partonic processes and find significant gains from using modern machine learning in event generators.15 pages, 6 figures, 2 tables; v3: updates incl. referee request
ELSA -- Enhanced latent spaces for improved collider simulations
No full textSimulations play a key role for inference in collider physics. We explore
various approaches for enhancing the precision of simulations using machine
learning, including interventions at the end of the simulation chain
(reweighting), at the beginning of the simulation chain (pre-processing), and
connections between the end and beginning (latent space refinement). To clearly
illustrate our approaches, we use W+jets matrix element surrogate simulations
based on normalizing flows as a prototypical example. First, weights in the
data space are derived using machine learning classifiers. Then, we pull back
the data-space weights to the latent space to produce unweighted examples and
employ the Latent Space Refinement (LASER) protocol using Hamiltonian Monte
Carlo. An alternative approach is an augmented normalizing flow, which allows
for different dimensions in the latent and target spaces. These methods are
studied for various pre-processing strategies, including a new and general
method for massive particles at hadron colliders that is a tweak on the
widely-used RAMBO-on-diet mapping. We find that modified simulations can
achieve sub-percent precision across a wide range of phase space.Comment: 17 pages, 9 figures, 2 tables, code and data at
https://github.com/ramonpeter/elsa, v2: journal versio
How to GAN : Novel simulation methods for the LHC
No full textVarious aspects of LHC simulations can be supplemented by generative networks. For event generation we show how a GAN can describe the full phase space structure of top-pair production including intermediate on-shell resonances and phase space bound- aries. In order to resolve these sharp peaking features, we introduce the maximum mean discrepancy. Additionally, the architecture can be extended in a straightforward manner to improve the network performance and to handle weighted events in the training data. Furthermore, we employ GANs to generate new events which are distributed according to the sum or difference of the input data. We first show with the help of a toy example how such a network can beat the statistical limitations of bin-wise subtraction methods. Afterwards we demonstrate how this network can subtract background events or describe collinear subtraction events in next-to-leading order calculations. Finally, we show how detector simulations can be inverted using GANs and INNs. They allow us to reconstruct parton level information from measured events. In detail, our results show how conditional generative networks can invert Monte Carlo simulations statistically. INNs even allow for a statistical interpretation of single-event unfolding and yield the possibility to unfold parton showering
Accurate Surrogate Amplitudes with Calibrated Uncertainties
No full textNeural networks for LHC physics have to be accurate, reliable, and
controlled. Using surrogate loop amplitudes as a use case, we first show how
activation functions can be systematically tested with KANs. For reliability
and control, we learn uncertainties together with the target amplitude over
phase space. Systematic uncertainties can be learned by a heteroscedastic loss,
but a comprehensive learned uncertainty requires Bayesian networks or repulsive
ensembles. We compute pull distributions to show to what level learned
uncertainties are calibrated correctly for cutting-edge precision surrogates
Like-Sign W-Boson Scattering at the LHC -- Approximations and Full Next-to-Leading-Order Predictions
No full textWe present a new calculation of next-to-leading-order corrections of the
strong and electroweak interactions to like-sign W-boson scattering at the
Large Hadron Collider, implemented in the Monte Carlo integrator Bonsay. The
calculation includes leptonic decays of the bosons. It comprises
the whole tower of next-to-leading-order contributions to the cross section,
which scale like , ,
, and in the strong and electroweak
couplings and . We present a detailed survey of
numerical results confirming the occurrence of large pure electroweak
corrections of the order of for integrated cross sections and even
larger corrections in high-energy tails of distributions. The electroweak
corrections account for the major part of the complete next-to-leading-order
correction, which amounts to in size, depending on the details of
the event selection chosen for analysing vector-boson-scattering. Moreover, we
compare the full next-to-leading-order corrections to approximate results based
on the neglect of contributions that are not enhanced by the vector-boson
scattering kinematics (VBS approximation) and on resonance expansions for the
-boson decays (double-pole approximation); the quality of this
approximation is good within for integrated cross sections and the
dominating parts of the differential distributions. Finally, for the
leading-order predictions, we construct different versions of effective
vector-boson approximations, which are based on cross-section contributions
that are enhanced by collinear emission of bosons off the
initial-state (anti)quarks; in line with previous findings in the literature,
it turns out that the approximative quality is rather limited for applications
at the LHC.Comment: 57 pages, 70 figures; version published in JHE
Full and approximated NLO predictions for like-sign W-boson scattering at the LHC
No full textWe report on a recent calculation of next-to-leading-order (NLO) QCD and electroweakcorrections to like-sign W-boson scattering at the Large Hadron Collider, including allpartonic channels and W-boson decays in the process.The calculation is implemented in the Monte Carlo integrator Bonsay andcomprises the full tower of NLO contributions of the orders, , , and .Our numerical results confirm and extend previous results, in particular the occurrenceof large purely electroweak corrections of the order of for integrated cross sections,which get even larger in distributions.We construct a "VBS approximation'' for the NLO prediction based onpartonic channels and gauge-invariant (sub)matrix elements potentially containing thevector-boson scattering (VBS) subprocess and on resonance expansions of the W decays.The VBS approximation reproduces the full NLO predictions within in the most important regions of phase space.Moreover, we discuss results from different versions of "effective vector-boson approximations''at leading order, based on the collinear emission of W bosons of incoming (anti)quarks.However, owing to the only mild collinear enhancement and the design of VBS analysis cuts,the quality of this approximation turns out to be only qualitative at the LHC.We report on a recent calculation of next-to-leading-order (NLO) QCD and electroweak corrections to like-sign W-boson scattering at the Large Hadron Collider, including all partonic channels and W-boson decays in the process . The calculation is implemented in the Monte Carlo integrator Bonsay and comprises the full tower of NLO contributions of the orders , , , and . Our numerical results confirm and extend previous results, in particular the occurrence of large purely electroweak corrections of the order of for integrated cross sections, which get even larger in distributions. We construct a "VBS approximation" for the NLO prediction based on partonic channels and gauge-invariant (sub)matrix elements potentially containing the vector-boson scattering (VBS) subprocess and on resonance expansions of the Wdecays. The VBS approximation reproduces the full NLO predictions within in the most important regions of phase space. Moreover, we discuss results from different versions of "effective vector-boson approximations" at leading order, based on the collinear emission of W bosons of incoming (anti)quarks. However, owing to the only mild collinear enhancement and the design of VBS analysis cuts, the quality of this approximation turns out to be only qualitative at the LHC
Differentiable MadNIS-Lite
No full textDifferentiable programming opens exciting new avenues in particle physics, also affecting future event generators. These new techniques boost the performance of current and planned MadGraph implementations. Combining phase-space mappings with a set of very small learnable flow elements, MADNIS-Lite, can improve the sampling efficiency while being physically interpretable. This defines a third sampling strategy, complementing VEGAS and the full MADNIS
Full and approximated NLO predictions for like-sign W-boson scattering at the LHC
No full textWe report on a recent calculation of next-to-leading-order (NLO) QCD and
electroweak corrections to like-sign W-boson scattering at the Large Hadron
Collider, including all partonic channels and W-boson decays in the process . The calculation is implemented in the
Monte Carlo integrator Bonsay and comprises the full tower of NLO contributions
of the orders , , ,
and . Our numerical results confirm and extend previous results, in
particular the occurrence of large purely electroweak corrections of the order
of for integrated cross sections, which get even larger in
distributions. We construct a "VBS approximation" for the NLO prediction based
on partonic channels and gauge-invariant (sub)matrix elements potentially
containing the vector-boson scattering (VBS) subprocess and on resonance
expansions of the Wdecays. The VBS approximation reproduces the full NLO
predictions within in the most important regions of phase space.
Moreover, we discuss results from different versions of "effective vector-boson
approximations" at leading order, based on the collinear emission of W bosons
of incoming (anti)quarks. However, owing to the only mild collinear enhancement
and the design of VBS analysis cuts, the quality of this approximation turns
out to be only qualitative at the LHC.Comment: 10 pages, latex, 8 figures, proceedings contribution to "Loops and
Legs in Quantum Field Theory (LL2024)", April 2024, Wittenberg, German
