183 Search Results

Abstract Simulations 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 andmore » 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 RamboOnDiet mapping. We find that modified simulations can achieve sub-percent precision across a wide range of phase space.« less

Many particle physics datasets like those generated at colliders are described by continuous coordinates (in contrast to grid points like in an image), respect a number of symmetries (like permutation invariance), and have a stochastic dimensionality. For this reason, standard deep generative models that produce images or at least a fixed set of features are limiting. We introduce a new neural network simulation based on a diffusion model that addresses these limitations named fast point cloud diffusion. We show that our approach can reproduce the complex properties of hadronic jets from proton-proton collisions with competitive precision to other recently proposedmore » models. Additionally, we use a procedure called progressive distillation to accelerate the generation time of our method, which is typically a significant challenge for diffusion models despite their state-of-the-art precision.« less

Precise reconstruction of top quark properties is a challenging task at the Large Hadron Collider due to combinatorial backgrounds and missing information. We introduce a physics-informed neural network architecture called the covariant particle transformer (CPT) for directly predicting the top quark kinematic properties from reconstructed final state objects. This approach is permutation invariant and partially Lorentz covariant and can account for a variable number of input objects. In contrast to previous machine learning-based reconstruction methods, CPT is able to predict top quark four-momenta regardless of the jet multiplicity in the event. Using simulations, we show that the CPT performs favorablymore » compared with other machine learning top quark reconstruction approaches.« less

Resonant anomaly detection is a promising framework for model-independent searches for new particles. Weakly supervised resonant anomaly detection methods compare data with a potential signal against a template of the Standard Model (SM) background inferred from sideband regions. We propose a means to generate this background template that uses a flow-based model to create a mapping between high-fidelity SM simulations and the data. The flow is trained in sideband regions with the signal region blinded, and the flow is conditioned on the resonant feature (mass) such that it can be interpolated into the signal region. To illustrate this approach, wemore » use simulated collisions from the Large Hadron Collider (LHC) Olympics dataset. We find that our flow-constructed background method has competitive sensitivity with other recent proposals and can therefore provide complementary information to improve future searches.« less

Hadronization is a critical step in the simulation of high-energy particle and nuclear physics experiments. As there is no first principles understanding of this process, physically-inspired hadronization models have a large number of parameters that are fit to data. Deep generative models are a natural replacement for classical techniques, since they are more flexible and may be able to improve the overall precision. Proof of principle studies have shown how to use neural networks to emulate specific hadronization when trained using the inputs and outputs of classical methods. However, these approaches will not work with data, where we do notmore » have a matching between observed hadrons and partons. In this paper, we develop a protocol for fitting a deep generative hadronization model in a realistic setting, where we only have access to a set of hadrons in data. Our approach uses a variation of a Generative Adversarial Network with a permutation invariant discriminator. We find that this setup is able to match the hadronization model in Herwig with multiple sets of parameters. This work represents a significant step forward in a longer term program to develop, train, and integrate machine learning-based hadronization models into parton shower Monte Carlo programs.« less

A search for heavy right-handed Majorana or Dirac neutrinos N_R and heavy right-handed gauge bosons W_R is performed in events with energetic electrons or muons, with the same or opposite electric charge, and energetic jets. The search is carried out separately for topologies of clearly separated final-state products (“resolved” channel) and topologies with boosted final states with hadronic and/or leptonic products partially overlapping and reconstructed as a large-radius jet (“boosted” channel). The events are selected from pp collision data at the LHC with an integrated luminosity of 139 fb^-1 collected by the ATLAS detector at $$\sqrt{s}=13$$ TeV. No significant deviationsmore » from the Standard Model predictions are observed. The results are interpreted within the theoretical framework of a left-right symmetric model, and lower limits are set on masses in the heavy right handed W_R boson and N_R plane. The excluded region extends to about m(W_R) = 6.4 TeV for both Majorana and Dirac N_R neutrinos at m(N_R) < 1 TeV. N_R with masses of less than 3.5 (3.6) TeV are excluded in the electron (muon) channel at m(W_R) = 4.8 TeV for the Majorana neutrinos, and lim its of m(N_R) up to 3.6 TeV for m(W_R) = 5.2 (5.0) TeV in the electron (muon) channel are set for the Dirac neutrinos. These constitute the most stringent exclusion limits to date for the model considered.« less

The inclusive top-quark-pair production cross section σ_{$$t\overline{t}$$} and its ratio to the Z-boson production cross section have been measured in proton–proton collisions at $$\sqrt{s}$$ = 13.6 TeV, using 29 fb^—1 of data collected in 2022 with the ATLAS experiment at the Large Hadron Collider. Using events with an opposite-charge electron-muon pair and b-tagged jets, and assuming Standard Model decays, the top-quark-pair production cross section is measured to be σ_{$$t\overline{t}$$} = 850± 3(stat.) ± 18(syst.) ± 20(lumi.) pb. The ratio of the $$t\overline{t}$$ and the Z-boson production cross sections is also measured, where the Z-boson contribution is determined for inclusive e⁺e^—more » and μ⁺μ^— events in a fiducial phase space. The relative uncertainty on the ratio is reduced compared to the $$t\overline{t}$$ cross section, thanks to the cancellation of several systematic uncertainties. The result for the ratio, R_{$$t\overline{t}/Z$$} = 1.145 ± 0.003(stat.) ± 0.021(syst.)±0.002(lumi.) is consistent with the Standard Model prediction using the PDF4LHC21 PDF set.« less

Measurements and searches performed with the ATLAS detector at the CERN LHC often involve signatures with one or more prompt leptons. Such analyses are subject to `fake/non-prompt' lepton backgrounds, where either a hadron or a lepton from a hadron decay or an electron from a photon conversion satisfies the prompt-lepton selection criteria. These backgrounds often arise within a hadronic jet because of particle decays in the showering process, particle misidentification or particle interactions with the detector material. As it is challenging to model these processes with high accuracy in simulation, their estimation typically uses data-driven methods. Three methods for carryingmore » out this estimation are described, along with their implementation in ATLAS and their performance.« less

A search for a new pseudoscalar a-boson produced in events with a top-quark pair, where the a-boson decays into a pair of muons, is performed using $$\sqrt{s}$$=13 TeV pp collision data collected with the ATLAS detector at the LHC, corresponding to an integrated luminosity of 139 fb^-1. The search targets the final state where only one top quark decays to an electron or muon, resulting in a signature with three leptons eμμ and μμμ. No significant excess of events above the Standard Model expectation is observed and upper limits are set on two signal models: $$pp→t\bar{t}a$$ and $$pp→t\bar{t}$$ with $t→H^±b,more » H^±→W^±a$, where $a→μμ$, in the mass ranges 15 GeV_a<72 GeV and 120 GeV≤m_H±≤160 GeV.« less

Searches for long-lived particles (LLPs) are among the most promising avenues for discovering physics beyond the Standard Model at the Large Hadron Collider (LHC). However, displaced signatures are notoriously difficult to identify due to their ability to evade standard object reconstruction strategies. In particular, the ATLAS track reconstruction applies strict pointing requirements which limit sensitivity to charged particles originating far from the primary interaction point. To recover efficiency for LLPs decaying within the tracking detector volume, the ATLAS Collaboration employs a dedicated large-radius tracking (LRT) pass with loosened pointing requirements. During Run 2 of the LHC, the LRT implementation producedmore » many incorrectly reconstructed tracks and was therefore only deployed in small subsets of events. In preparation for LHC Run 3, ATLAS has significantly improved both standard and large-radius track reconstruction performance, allowing for LRT to run in all events. This development greatly expands the potential phase-space of LLP searches and streamlines LLP analysis workflows. This paper will highlight the above achievement and report on the readiness of the ATLAS detector for track-based LLP searches in Run 3.« less