arXiv subjects

Timothy Hoffman

Publications and source records attributed to Timothy Hoffman.

5 recordsLinked to original sources

arXiv2023

19 Parameters Is All You Need: Tiny Neural Networks for Particle Physics

As particle accelerators increase their collision rates, and deep learning solutions prove their viability, there is a growing need for lightweight and fast neural network architectures for low-latency tasks such as triggering. We examine the potential of one recent Lorentz- and permutation-symmetric architecture, PELICAN, and present its instances with as few as 19 trainable parameters that outperform generic architectures with tens of thousands of parameters when compared on the binary classification task of top quark jet tagging.

hep-ph↗

arXiv2023

Explainable Equivariant Neural Networks for Particle Physics: PELICAN

PELICAN is a novel permutation equivariant and Lorentz invariant or covariant aggregator network designed to overcome common limitations found in architectures applied to particle physics problems. Compared to many approaches that use non-specialized architectures that neglect underlying physics principles and require very large numbers of parameters, PELICAN employs a fundamentally symmetry group-based architecture that demonstrates benefits in terms of reduced complexity, increased interpretability, and raw performance. We present a comprehensive study of the PELICAN algorithm architecture in the context of both tagging (classification) and reconstructing (regression) Lorentz-boosted top quarks, including the difficult task of specifically identifying and measuring the $W$-boson inside the dense environment of the Lorentz-boosted top-quark hadronic final state. We also extend the application of PELICAN to the tasks of identifying quark-initiated vs.~gluon-initiated jets, and a multi-class identification across five separate target categories of jets. When tested on the standard task of Lorentz-boosted top-quark tagging, PELICAN outperforms existing competitors with much lower model complexity and high sample efficiency. On the less common and more complex task of 4-momentum regression, PELICAN also outperforms hand-crafted, non-machine learning algorithms. We discuss the implications of symmetry-restricted architectures for the wider field of machine learning for physics.

hep-ph↗

arXiv2022

PELICAN: Permutation Equivariant and Lorentz Invariant or Covariant Aggregator Network for Particle Physics

Many current approaches to machine learning in particle physics use generic architectures that require large numbers of parameters and disregard underlying physics principles, limiting their applicability as scientific modeling tools. In this work, we present a machine learning architecture that uses a set of inputs maximally reduced with respect to the full 6-dimensional Lorentz symmetry, and is fully permutation-equivariant throughout. We study the application of this network architecture to the standard task of top quark tagging and show that the resulting network outperforms all existing competitors despite much lower model complexity. In addition, we present a Lorentz-covariant variant of the same network applied to a 4-momentum regression task.

hep-ph↗

arXiv2017

Partially Acoustic Dark Matter Cosmology and Cosmological Constraints

Observations of the cosmic microwave background (CMB) together with weak lensing measurements of the clustering of large scale cosmological structures and local measurements of the Hubble constant pose a challenge to the standard $\Lambda$CDM cosmological model. On one side CMB observations imply a Hubble constant that is lower than local measurements and an amplitude of the lensing signal that is higher than direct measurements from weak lensing surveys. We investigate a way of relieving these tensions by adding dark radiation tightly coupled to an acoustic part of the dark matter sector and compare it to massive neutrino solutions. While these models offer a way of separately relieving the Hubble and weak lensing tensions they are prevented from fully accommodating both at the same time since the CMB requires additional cold dark matter when adding acoustic dark matter or massive neutrinos to preserve the same sharpness of the acoustic peaks which counteracts the desired growth suppression.

astro-ph.CO↗

arXiv2017

Ultra-low-loss Polaritons in Isotopically Pure Materials: A New Approach

Conventional optical components are limited to size-scales much larger than the wavelength of light, as changes in the amplitude, phase and polarization of the electromagnetic fields are accrued gradually along an optical path. However, advances in nanophotonics have produced ultra-thin, co-called "flat" optical components that beget abrupt changes in these properties over distances significantly shorter than the free space wavelength. While high optical losses still plague many approaches, phonon polariton (PhP) materials have demonstrated long lifetimes for sub-diffractional modes in comparison to plasmon-polariton-based nanophotonics. We experimentally observe a three-fold improvement in polariton lifetime through isotopic enrichment of hexagonal boron nitride (hBN). Commensurate increases in the polariton propagation length are demonstrated via direct imaging of polaritonic standing waves by means of infrared nano-optics. Our results provide the foundation for a materials-growth-directed approach towards realizing the loss control necessary for the development of PhP-based nanophotonic devices.

cond-mat.mtrl-sci↗