arXiv subjects

Tilman Plehn

Publications and source records attributed to Tilman Plehn.

At least 73 records · Page 4Linked to original sources

arXiv2024

The Landscape of Unfolding with Machine Learning

Recent innovations from machine learning allow for data unfolding, without binning and including correlations across many dimensions. We describe a set of known, upgraded, and new methods for ML-based unfolding. The performance of these approaches are evaluated on the same two datasets. We find that all techniques are capable of accurately reproducing the particle-level spectra across complex observables. Given that these approaches are conceptually diverse, they offer an exciting toolkit for a new class of measurements that can probe the Standard Model with an unprecedented level of detail and may enable sensitivity to new phenomena.

hep-ph↗

arXiv2024

PINNferring the Hubble Function with Uncertainties

The Hubble function characterizes a given Friedmann-Robertson-Walker spacetime and can be related to the densities of the cosmological fluids and their equations of state. We show how physics-informed neural networks (PINNs) emulate this dynamical system and provide fast predictions of the luminosity distance for a given choice of densities and equations of state, as needed for the analysis of supernova data. We use this emulator to perform a model-independent and parameter-free reconstruction of the Hubble function on the basis of supernova data. As part of this study, we develop and validate an uncertainty treatment for PINNs using a heteroscedastic loss and repulsive ensembles.

astro-ph.CO↗

arXiv2024

To Profile or To Marginalize -- A SMEFT Case Study

Global SMEFT analyses have become a key interpretation framework for LHC physics, quantifying how well a large set of kinematic measurements agrees with the Standard Model. This agreement is encoded in measured Wilson coefficients and their uncertainties. A technical challenge of global analyses are correlations. We compare, for the first time, results from a profile likelihood and a Bayesian marginalization for a given data set with a comprehensive uncertainty treatment. Using the validated Bayesian framework we analyse a series of new kinematic measurements. For the updated dataset we find and explain differences between the marginalization and profile likelihood treatments.

hep-ph↗

arXiv2023

How to Understand Limitations of Generative Networks

Well-trained classifiers and their complete weight distributions provide us with a well-motivated and practicable method to test generative networks in particle physics. We illustrate their benefits for distribution-shifted jets, calorimeter showers, and reconstruction-level events. In all cases, the classifier weights make for a powerful test of the generative network, identify potential problems in the density estimation, relate them to the underlying physics, and tie in with a comprehensive precision and uncertainty treatment for generative networks.

hep-ph↗

arXiv2023

What's Anomalous in LHC Jets?

Searches for anomalies are a significant motivation for the LHC and help define key analysis steps, including triggers. We discuss specific examples how LHC anomalies can be defined through probability density estimates, evaluated in a physics space or in an appropriate neural network latent space, and discuss the model-dependence in choosing an appropriate data parameterisation. We illustrate this for classical k-means clustering, a Dirichlet variational autoencoder, and invertible neural networks. For two especially challenging scenarios of jets from a dark sector we evaluate the strengths and limitations of each method.

hep-ph↗

arXiv2023

MadNIS -- Neural Multi-Channel Importance Sampling

Theory predictions for the LHC require precise numerical phase-space integration and generation of unweighted events. We combine machine-learned multi-channel weights with a normalizing flow for importance sampling, to improve classical methods for numerical integration. We develop an efficient bi-directional setup based on an invertible network, combining online and buffered training for potentially expensive integrands. We illustrate our method for the Drell-Yan process with an additional narrow resonance.

hep-ph↗

arXiv2023

Returning CP-Observables to The Frames They Belong

Optimal kinematic observables are often defined in specific frames and then approximated at the reconstruction level. We show how multi-dimensional unfolding methods allow us to reconstruct these observables in their proper rest frame and in a probabilistically faithful way. We illustrate our approach with a measurement of a CP-phase in the top Yukawa coupling. Our method makes use of key advantages of generative unfolding, but as a constructed observable it fits into standard LHC analysis frameworks.

hep-ph↗

arXiv2023

A Normalized Autoencoder for LHC Triggers

Autoencoders are an effective analysis tool for the LHC, as they represent one of its main goal of finding physics beyond the Standard Model. The key challenge is that out-of-distribution anomaly searches based on the compressibility of features do not apply to the LHC, while existing density-based searches lack performance. We present the first autoencoder which identifies anomalous jets symmetrically in the directions of higher and lower complexity. The normalized autoencoder combines a standard bottleneck architecture with a well-defined probabilistic description. It works better than all available autoencoders for top vs QCD jets and reliably identifies different dark-jet signals.

hep-ph↗

arXiv2023

Two Invertible Networks for the Matrix Element Method

The matrix element method is widely considered the ultimate LHC inference tool for small event numbers. We show how a combination of two conditional generative neural networks encodes the QCD radiation and detector effects without any simplifying assumptions, while keeping the computation of likelihoods for individual events numerically efficient. We illustrate our approach for the CP-violating phase of the top Yukawa coupling in associated Higgs and single-top production. Currently, the limiting factor for the precision of our approach is jet combinatorics.

hep-ph↗

arXiv2023

Targeting Multi-Loop Integrals with Neural Networks

Numerical evaluations of Feynman integrals often proceed via a deformation of the integration contour into the complex plane. While valid contours are easy to construct, the numerical precision for a multi-loop integral can depend critically on the chosen contour. We present methods to optimize this contour using a combination of optimized, global complex shifts and a normalizing flow. They can lead to a significant gain in precision.

hep-ph↗

arXiv2023

Statistical Patterns of Theory Uncertainties

A comprehensive uncertainty estimation is vital for the precision program of the LHC. While experimental uncertainties are often described by stochastic processes and well-defined nuisance parameters, theoretical uncertainties lack such a description. We study uncertainty estimates for cross-section predictions based on scale variations across a large set of processes. We find patterns similar to a stochastic origin, with accurate uncertainties for processes mediated by the strong force, but a systematic underestimate for electroweak processes. We propose an improved scheme, based on the scale variation of reference processes, which reduces outliers in the mapping from leading order to next-to-leading-order in perturbation theory.

hep-ph↗

arXiv2023

Cornering Extended Starobinsky Inflation with CMB and SKA

Starobinsky inflation is an attractive, fundamental model to explain the Planck measurements, and its higher-order extension may allow us to probe quantum gravity effects. We show that future CMB data combined with the 21cm intensity map from SKA will meaningfully probe such an extended Starobinsky model. A combined analysis will provide a precise measurement and intriguing insight into inflationary dynamics, even accounting for correlations with astrophysical parameters.

astro-ph.CO↗

arXiv2023

Back to the Formula -- LHC Edition

While neural networks offer an attractive way to numerically encode functions, actual formulas remain the language of theoretical particle physics. We show how symbolic regression trained on matrix-element information provides, for instance, optimal LHC observables in an easily interpretable form. We introduce the method using the effect of a dimension-6 coefficient on associated ZH production. We then validate it for the known case of CP-violation in weak-boson-fusion Higgs production, including detector effects.

hep-ph↗

arXiv2023

Calomplification -- The Power of Generative Calorimeter Models

Motivated by the high computational costs of classical simulations, machine-learned generative models can be extremely useful in particle physics and elsewhere. They become especially attractive when surrogate models can efficiently learn the underlying distribution, such that a generated sample outperforms a training sample of limited size. This kind of GANplification has been observed for simple Gaussian models. We show the same effect for a physics simulation, specifically photon showers in an electromagnetic calorimeter.

hep-ph↗

arXiv2022

Machine Learning and LHC Event Generation

First-principle simulations are at the heart of the high-energy physics research program. They link the vast data output of multi-purpose detectors with fundamental theory predictions and interpretation. This review illustrates a wide range of applications of modern machine learning to event generation and simulation-based inference, including conceptional developments driven by the specific requirements of particle physics. New ideas and tools developed at the interface of particle physics and machine learning will improve the speed and precision of forward simulations, handle the complexity of collision data, and enhance inference as an inverse simulation problem.

hep-ph↗

arXiv2022

Performance versus Resilience in Modern Quark-Gluon Tagging

Discriminating quark-like from gluon-like jets is, in many ways, a key challenge for many LHC analyses. First, we use a known difference in Pythia and Herwig simulations to show how decorrelated taggers would break down when the most distinctive feature is aligned with theory uncertainties. We propose conditional training on interpolated samples, combined with a controlled Bayesian network, as a more resilient framework. The interpolation parameter can be used to optimize the training evaluated on a calibration dataset, and to test the stability of this optimization. The interpolated training might also be useful to track generalization errors when training networks on simulation.

hep-ph↗

arXiv2022

Generative Networks for Precision Enthusiasts

Generative networks are opening new avenues in fast event generation for the LHC. We show how generative flow networks can reach percent-level precision for kinematic distributions, how they can be trained jointly with a discriminator, and how this discriminator improves the generation. Our joint training relies on a novel coupling of the two networks which does not require a Nash equilibrium. We then estimate the generation uncertainties through a Bayesian network setup and through conditional data augmentation, while the discriminator ensures that there are no systematic inconsistencies compared to the training data.

hep-ph↗

arXiv2022

Hazma Meets HERWIG4DM: Precision Gamma-Ray, Neutrino, and Positron Spectra for Light Dark Matter

We present a new open-source package, Hazma 2, that computes accurate spectra relevant for indirect dark matter searches for photon, neutrino, and positron production from vector-mediated dark matter annihilation and for spin-one dark matter decay. The tool bridges across the regimes of validity of two state of the art codes: Hazma 1, which provides an accurate description below hadronic resonances up to center-of-mass energies around 250 MeV, and HERWIG4DM, which is based on vector meson dominance and measured form factors, and accurate well into the few GeV range. The applicability of the combined code extends to approximately 1.5 GeV, above which the number of final state hadrons off of which we individually compute the photon, neutrino, and positron yield grows exceedingly rapidly. We provide example branching ratios, particle spectra and conservative observational constraints from existing gamma-ray data for the well-motivated cases of decaying dark photon dark matter and vector-mediated fermionic dark matter annihilation. Finally, we compare our results to other existing codes at the boundaries of their respective ranges of applicability. Hazma 2 is freely available on GitHub.

hep-ph↗