Publications

Order by: publication date number of citations

• Towards a data-driven model of hadronization using normalizing flows

  Christian Bierlich, Phil Ilten, Tony Menzo, Stephen Mrenna, Manuel Szewc, Michael K. Wilkinson, Ahmed Youssef, Jure Zupan

  SciPost Phys. 17, 045 (2024) · published 12 August 2024 | Toggle abstract · pdf

  We introduce a model of hadronization based on invertible neural networks that faithfully reproduces a simplified version of the Lund string model for meson hadronization. Additionally, we introduce a new training method for normalizing flows, termed MAGIC, that improves the agreement between simulated and experimental distributions of high-level (macroscopic) observables by adjusting single-emission (microscopic) dynamics. Our results constitute an important step toward realizing a machine-learning based model of hadronization that utilizes experimental data during training. Finally, we demonstrate how a Bayesian extension to this normalizing-flow architecture can be used to provide analysis of statistical and modeling uncertainties on the generated observable distributions.

  1 citation

• Codebase release 1.1 for GAPS

  Michael H. Seymour, Siddharth Sule

  SciPost Phys. Codebases 33-r1.1 (2024) · published 12 August 2024 | Toggle abstract · src

  The Single Instruction, Multiple Thread (SIMT) paradigm of GPU programming does not support the branching nature of a parton shower algorithm by definition. However, modern GPUs are designed to schedule threads with diverging processes independently, allowing them to handle such branches. With regular thread synchronisation and careful treatment of the individual steps, one can simulate a parton shower on a GPU. We present a Sudakov veto algorithm designed to simulate parton branching on multiple events in parallel. We also release a CUDA C++ program that generates matrix elements, showers partons and computes jet rates and event shapes for LEP at 91.2 GeV on a GPU. To benchmark its performance, we also provide a near-identical C++ program designed to simulate events serially on a CPU. While the consequences of branching are not absent, we demonstrate that a GPU can provide the throughput of a many-core CPU. As an example, we show that the time taken to shower $10^6$ events on one NVIDIA TESLA V100 GPU is equivalent to that of 295 Intel Xeon E5-2620 CPU cores.

  1 citation

• An algorithm to parallelise parton showers on a GPU

  Michael H. Seymour, Siddharth Sule

  SciPost Phys. Codebases 33 (2024) · published 12 August 2024 | Toggle abstract · pdf

  The Single Instruction, Multiple Thread (SIMT) paradigm of GPU programming does not support the branching nature of a parton shower algorithm by definition. However, modern GPUs are designed to schedule threads with diverging processes independently, allowing them to handle such branches. With regular thread synchronisation and careful treatment of the individual steps, one can simulate a parton shower on a GPU. We present a Sudakov veto algorithm designed to simulate parton branching on multiple events in parallel. We also release a CUDA C++ program that generates matrix elements, showers partons and computes jet rates and event shapes for LEP at 91.2 GeV on a GPU. To benchmark its performance, we also provide a near-identical C++ program designed to simulate events serially on a CPU. While the consequences of branching are not absent, we demonstrate that a GPU can provide the throughput of a many-core CPU. As an example, we show that the time taken to shower $10^6$ events on one NVIDIA TESLA V100 GPU is equivalent to that of 295 Intel Xeon E5-2620 CPU cores.

  1 citation

• “Reborn in Guate”: Making Resource Frontiers in Asylum in Guatemala’s Northern Petén

  Julia Morris

  Mig. Pol. 3, 003 (2024) · published 12 August 2024 | Toggle abstract · pdf

  The last decade and a half have seen a dramatic increase in the outsourcing and offshoring of asylum processing and resettlement to countries in the Global South. This article advances a new theoretical framework to examine the surge in new asylum regimes worldwide. Drawing on ethnographic fieldwork in several externalised asylum sites and specifically in Guatemala, it looks at these recent developments through the lens of ‘resource frontiers.’ Merging critical political ecological approaches on resource frontiers with research on border externalisation, I argue that ‘asylum frontiers’ are the social spaces connected to the exploration and development of a resource sector that extracts value from people on the move. I centre my analysis on the US-driven development of an asylum regime in Guatemala’s northern Petén region. I consider the specificities of Guatemala’s emerging asylum frontier, detailing how this arrangement sits with the country’s own histories of asylum and enforced return. In doing so, I show how different political actors – migrants, Indigenous Mayan refugees, and deported Guatemalans – ‘live with’ these frontier economies.

• Localization engineering by resonant driving in dissipative polariton arrays

  Gonzalo Usaj

  SciPost Phys. Core 7, 052 (2024) · published 9 August 2024 | Toggle abstract · pdf

  Arrays of microcavity polaritons are very versatile systems that allow for broad possibilities for the engineering of multi-orbital lattice geometries using different state preparation schemes. One of these schemes, spatially modulated resonant driving, can be used, for instance, to selectively localize the polariton field within the particular region of the lattice enclosed by the driving laser. Both the frequency and the spatial amplitude distribution (module and phase) of the driven laser field are important and serve as a knob to control the leakage outside that region and hence the extend of the spatial localization. Here, we analyse both the linear and nonlinear regimes using the lattice Green function formalism that is particularly suitable for the case of polariton arrays described in a tight-binding approximation. We identify the conditions for the laser induced localization to occur on arbitrary lattice's geometries and discuss some experimentally relevant cases. We find that the polariton-polariton interaction leads to a frequency shift of the optimal localization condition that could be used to further control it.

• Open string stub as an auxiliary string field

  Harold Erbin, Atakan Hilmi Firat

  SciPost Phys. 17, 044 (2024) · published 9 August 2024 | Toggle abstract · pdf

  Witten's open string field theory with a generalized version of stubs is reformulated as a cubic string field theory using an auxiliary string field. The gauge symmetries and equations of motion as well as the associative algebra of the resulting theory are investigated. Integrating out either the original or auxiliary field is shown to recover the conventional cubic theory. Our analysis demonstrates that deformations due to the stubs can be described as a homotopy transfer purely in the context of strong deformation retract. We also discuss to what extent the vertex regions resulting from stubs provide a model for the elementary interactions of closed string field theory.

  1 citation

• Precision magnetometry exploiting excited state quantum phase transitions

  Qian Wang, Ugo Marzolino

  SciPost Phys. 17, 043 (2024) · published 9 August 2024 | Toggle abstract · pdf

  Critical behaviour in phase transitions is a resource for enhanced precision metrology. The reason is that the function, known as Fisher information, is superextensive at critical points, and, at the same time, quantifies performances of metrological protocols. Therefore, preparing metrological probes at phase transitions provides enhanced precision in measuring the transition control parameter. We focus on the Lipkin-Meshkov-Glick model that exhibits excited state quantum phase transitions at different magnetic fields. Resting on the model spectral properties, we show broad peaks of the Fisher information, and propose efficient schemes for precision magnetometry. The Lipkin-Meshkov-Glick model was first introduced for superconductivity and for nuclear systems, and recently realised in several condensed matter platforms. The above metrological schemes can be also exploited to measure microscopic properties of systems able to simulate the Lipkin-Meshkov-Glick model.

• Hydrodynamic fluctuations and topological susceptibility in chiral magnetohydrodynamics

  Arpit Das, Nabil Iqbal, Napat Poovuttikul

  SciPost Phys. 17, 042 (2024) · published 9 August 2024 | Toggle abstract · pdf

  Chiral magnetohydrodynamics is devoted to understanding the late-time and long-distance behavior of a system with an Adler-Bell-Jackiw anomaly at finite temperatures. The non-conservation of the axial charge is determined by the topological density $\vec{E} \cdot \vec{B}$; in a classical hydrodynamic description this decay rate can be suppressed by tuning the background magnetic field to zero. However it is in principle possible for thermal fluctuations of $\vec{E} \cdot \vec{B}$ to result in a non-conservation of the charge even at vanishing $B$-field; this would invalidate the classical hydrodynamic effective theory. We investigate this by computing the real-time susceptibility of the topological density at one-loop level in magnetohydrodynamic fluctuations, relating its low-frequency limit to the decay rate of the axial charge. We find that the frequency-dependence of this susceptibility is sufficiently soft as to leave the axial decay rate unaffected, validating the classical hydrodynamic description. We show that the susceptibility contains non-analytic frequency-dependence which is universally determined by hydrodynamic data. We comment briefly on possible connections to the recent formulation of the ABJ anomaly in terms of non-invertible symmetry.

• Quantum work capacitances: Ultimate limits for energy extraction on noisy quantum batteries

  Salvatore Tirone, Raffaele Salvia, Stefano Chessa, Vittorio Giovannetti

  SciPost Phys. 17, 041 (2024) · published 9 August 2024 | Toggle abstract · pdf

  We present a theoretical analysis of the energy recovery efficiency for quantum batteries composed of many identical quantum cells undergoing noise. While the possibility of using quantum effects to speed up the charging processes of batteries have been vastly investigated, In order to traslate these ideas into working devices it is crucial to assess the stability of the storage phase in the quantum battery elements when they are in contact with environmental noise. In this work we formalize this problem introducing a series of operationally well defined figures of merit (the work capacitances and the Maximal Asymptotic Work/Energy Ratios) which gauge the highest efficiency one can attain in recovering useful energy from quantum battery models that are formed by large collections of identical and independent elements (quantum cells or q-cells). Explicit evaluations of such quantities are presented for the case where the energy storing system undergoes through dephasing and depolarizing noise.

• Dense Hopfield networks in the teacher-student setting

  Robin Thériault, Daniele Tantari

  SciPost Phys. 17, 040 (2024) · published 8 August 2024 | Toggle abstract · pdf

  Dense Hopfield networks with $p$-body interactions are known for their feature to prototype transition and adversarial robustness. However, theoretical studies have been mostly concerned with their storage capacity. We derive the phase diagram of pattern retrieval in the teacher-student setting of $p$-body networks, finding ferromagnetic phases reminiscent of the prototype and feature learning regimes. On the Nishimori line, we find the critical amount of data necessary for pattern retrieval, and we show that the corresponding ferromagnetic transition coincides with the paramagnetic to spin-glass transition of $p$-body networks with random memories. Outside of the Nishimori line, we find that the student can tolerate extensive noise when it has a larger $p$ than the teacher. We derive a formula for the adversarial robustness of such a student at zero temperature, corroborating the positive correlation between number of parameters and robustness in large neural networks. Our model also clarifies why the prototype phase of $p$-body networks is adversarially robust.