arXiv subjects

Daniel Murnane

Publications and source records attributed to Daniel Murnane.

26 records · Page 2Linked to original sources

arXiv2021

Graph Neural Network for Object Reconstruction in Liquid Argon Time Projection Chambers

This paper presents a graph neural network (GNN) technique for low-level reconstruction of neutrino interactions in a Liquid Argon Time Projection Chamber (LArTPC). GNNs are still a relatively novel technique, and have shown great promise for similar reconstruction tasks in the LHC. In this paper, a multihead attention message passing network is used to classify the relationship between detector hits by labelling graph edges, determining whether hits were produced by the same underlying particle, and if so, the particle type. The trained model is 84% accurate overall, and performs best on the EM shower and muon track classes. The model's strengths and weaknesses are discussed, and plans for developing this technique further are summarised.

hep-ex↗

arXiv2021

Convergent Bayesian Global Fits of 4D Composite Higgs Models

Models in which the Higgs boson is a composite pseudo-Nambu-Goldstone boson offer attractive solutions to the Higgs mass naturalness problem. We consider three such models based on the minimal $SO(5) \rightarrow SO(4)$ symmetry breaking pattern, and perform convergent global fits on the models under a Bayesian framework in order to find the regions of their parameter spaces that best fit a wide range of constraints, including recent Higgs measurements. We use a novel technique to analyse the fine-tuning of the models, quantifying the tuning as the Kullback-Leibler divergence from the prior to the posterior probability on the parameter space. Each model is found to be able to satisfy all constraints at the $3σ$ level simultaneously. As a by-product of the fits, we analyse the collider phenomenology of our models in these viable regions. In two of the three models, we find that the $g g \rightarrow H \rightarrow γγ$ cross section is less than ${\sim}90$% that predicted by the SM, which is already in slight tension with experiment and could potentially be ruled out in the future high-luminosity run of the LHC. In addition, the lightest fermions $F$ arising from the new strong dynamics in these models are seen in general to lie above ${\sim}1.1$ TeV, with the $F \rightarrow tW^{+}$ and $F \rightarrow \bar{b}W^{+}$ decays offering particularly promising channels for probing these models in future collider searches.

hep-ph↗

arXiv2020

Minimal 4D Composite Higgs Models Under Current LHC Constraints

We present preliminary results of the first convergent global fits of several minimal composite Higgs models. Our fits are performed using the differential evolution optimisation package $\texttt{Diver}$. A variety of physical constraints are taken into account, including a wide range of exclusion bounds on heavy resonance production from Run 2 of the LHC. As a by-product of the fits, we analyse the collider phenomenology of the lightest new up-type and down-type resonances in the viable regions of our models, finding some low-mass resonances that can be probed in future collider searches.

hep-ph↗

arXiv2020

Track Seeding and Labelling with Embedded-space Graph Neural Networks

To address the unprecedented scale of HL-LHC data, the Exa.TrkX project is investigating a variety of machine learning approaches to particle track reconstruction. The most promising of these solutions, graph neural networks (GNN), process the event as a graph that connects track measurements (detector hits corresponding to nodes) with candidate line segments between the hits (corresponding to edges). Detector information can be associated with nodes and edges, enabling a GNN to propagate the embedded parameters around the graph and predict node-, edge- and graph-level observables. Previously, message-passing GNNs have shown success in predicting doublet likelihood, and we here report updates on the state-of-the-art architectures for this task. In addition, the Exa.TrkX project has investigated innovations in both graph construction, and embedded representations, in an effort to achieve fully learned end-to-end track finding. Hence, we present a suite of extensions to the original model, with encouraging results for hitgraph classification. In addition, we explore increased performance by constructing graphs from learned representations which contain non-linear metric structure, allowing for efficient clustering and neighborhood queries of data points. We demonstrate how this framework fits in with both traditional clustering pipelines, and GNN approaches. The embedded graphs feed into high-accuracy doublet and triplet classifiers, or can be used as an end-to-end track classifier by clustering in an embedded space. A set of post-processing methods improve performance with knowledge of the detector physics. Finally, we present numerical results on the TrackML particle tracking challenge dataset, where our framework shows favorable results in both seeding and track finding.

physics.ins-det↗

arXiv2020

Graph Neural Networks for Particle Reconstruction in High Energy Physics detectors

Pattern recognition problems in high energy physics are notably different from traditional machine learning applications in computer vision. Reconstruction algorithms identify and measure the kinematic properties of particles produced in high energy collisions and recorded with complex detector systems. Two critical applications are the reconstruction of charged particle trajectories in tracking detectors and the reconstruction of particle showers in calorimeters. These two problems have unique challenges and characteristics, but both have high dimensionality, high degree of sparsity, and complex geometric layouts. Graph Neural Networks (GNNs) are a relatively new class of deep learning architectures which can deal with such data effectively, allowing scientists to incorporate domain knowledge in a graph structure and learn powerful representations leveraging that structure to identify patterns of interest. In this work we demonstrate the applicability of GNNs to these two diverse particle reconstruction problems.

physics.ins-det↗

arXiv2019

ColliderBit: a GAMBIT module for the calculation of high-energy collider observables and likelihoods

We describe ColliderBit, a new code for the calculation of high energy collider observables in theories of physics beyond the Standard Model (BSM). ColliderBit features a generic interface to BSM models, a unique parallelised Monte Carlo event generation scheme suitable for large-scale supercomputer applications, and a number of LHC analyses, covering a reasonable range of the BSM signatures currently sought by ATLAS and CMS. ColliderBit also calculates likelihoods for Higgs sector observables, and LEP searches for BSM particles. These features are provided by a combination of new code unique to ColliderBit, and interfaces to existing state-of-the-art public codes. ColliderBit is both an important part of the GAMBIT framework for BSM inference, and a standalone tool for efficiently applying collider constraints to theories of new physics.

hep-ph↗

arXiv2019

Exploring Fine-tuning of the Next-to-Minimal Composite Higgs Model

We perform a detailed study of the fine-tuning of the two-site, 4D, Next-to-Minimal Composite Higgs Model (NMCHM), based on the global symmetry breaking pattern $SO(6)\rightarrow SO(5)$. Using our previously-defined fine-tuning measure that correctly combines the effect of multiple sources of fine-tuning, we quantify the fine-tuning that is expected to result from future collider measurements of the Standard Model-like Higgs branching ratios, in addition to null searches for the new resonances in the model. We also perform a detailed comparison with the Minimal Composite Higgs Model, finding that there is in general little difference between the fine-tuning expected in the two scenarios, even after measurements at a high-luminosity, 1 TeV linear collider. Finally, we briefly consider the relationship between fine-tuning and the ability of the extra scalar in the NMCHM model to act as a dark matter candidate, finding that the realisation of a $Z_2$ symmetry that stabilises the scalar is amongst the most natural regions in the parameter space, regardless of future collider measurements.

hep-ph↗

arXiv2017

Constraining fine tuning in Composite Higgs Models with partially composite leptons

Minimal Composite Higgs Models (MCHM) have long provided a solution to the hierarchy problem of the Standard Model, yet suffer from various sources of fine tuning that are becoming increasingly problematic with the lack of new physics observations at the LHC. We develop a new fine tuning measure that accurately counts each contribution to fine tuning (single, double, triple, etc) that can occur in a theory with $n_p$ parameters, that must reproduce $n_o$ observables. We then use a novel scanning procedure to perform a comprehensive study of three different two-site, 4D, $SO(5)\rightarrow SO(4)$ MCHMs with all third generation fermions included, distinguished by the choice of the lepton embeddings. These are the MCHM$^{\textbf{5-5-5}}_{\textbf{5-5-5}}$, MCHM$^{\textbf{5-5-5}}_{\textbf{14-14-10}}$ and MCHM$^{\textbf{5-5-5}}_{\textbf{14-1-10}}$, where MCHM$^{q-t-b}_{l-τ-ν}$ has the lepton doublet partner in representation $l$, tau partner in representation $τ$, and so on. We find that embedding at least one massive lepton in the symmetric $\textbf{14}$ of $SO(5)$ reduces the tuning for the case of low top partner masses (in line with previous results), but that this is balanced against the increased complexity of the model when one properly accounts for all sources of fine tuning. We study both the current relative fine-tuning of each scenario, and the future prospects. Noting that the different scenarios behave differently with respect to future improvements in collider measurements, we find that the MCHM$^{\textbf{5-5-5}}_{\textbf{14-1-10}}$ enjoys a relatively low increase in fine tuning even for a future lower bound on the top partner masses of 3.4 TeV (or equivalently a maximum Higgs-fermion or Higgs-gluon coupling deviation of 2%).

hep-ph↗