A5074395849- Particle physics theoretical and experimental studies
- High-Energy Particle Collisions Research
- Quantum Chromodynamics and Particle Interactions
- Particle Detector Development and Performance
- Medical Imaging Techniques and Applications
- Bayesian Methods and Mixture Models
- Radiation Detection and Scintillator Technologies
- Quantum Mechanics and Applications
University of Göttingen
2020-2023
We present a novel approach for the integration of scattering cross sections and generation partonic event samples in high-energy physics. propose an importance sampling technique capable overcoming typical deficiencies existing approaches by incorporating neural networks. The method guarantees full phase space coverage exact reproduction desired target distribution, our case given squared transition matrix element. study performance algorithm few representative examples, including top-quark...
First-principle simulations are at the heart of high-energy physics research program. They link vast data output multi-purpose detectors with fundamental theory predictions and interpretation. This review illustrates a wide range applications modern machine learning to event generation simulation-based inference, including conceptional developments driven by specific requirements particle physics. New ideas tools developed interface will improve speed precision forward simulations, handle...
The generation of unit-weight events for complex scattering processes presents a severe challenge to modern Monte Carlo event generators. Even when using sophisticated phase-space sampling techniques adapted the underlying transition matrix elements, efficiency generating from weighted samples can become limiting factor in practical applications. Here we present novel two-staged unweighting procedure that makes use neural-network surrogate full weight. algorithm significantly accelerate...
In this article we combine a recently proposed method for factorisation-aware matrix element surrogates with an unbiased unweighting algorithm. We show that employing sophisticated neural network emulation of QCD multijet elements based on dipole factorisation can lead to drastic acceleration unweighted event generation. train networks selection partonic channels contributing at the tree-level Z+4,5 <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"...
First-principle simulations are at the heart of high-energy physics research program. They link vast data output multi-purpose detectors with fundamental theory predictions and interpretation. This review illustrates a wide range applications modern machine learning to event generation simulation-based inference, including conceptional developments driven by specific requirements particle physics. New ideas tools developed interface will improve speed precision forward simulations, handle...
Abstract We present the first application of a Nested Sampling algorithm to explore high-dimensional phase space particle collision events. describe adaptation algorithm, designed perform Bayesian inference computations, integration partonic scattering cross sections and generation individual events distributed according corresponding squared matrix element. As concrete example we consider gluon processes into 3-, 4- 5-gluon final states compare performance with established sampling...
Abstract Modern machine learning methods offer great potential for increasing the efficiency of Monte Carlo event generators. We present latest developments in context SHERPA generation framework. These include phase space sampling amended by normalizing flows and a new unweighting procedure based on neural-network surrogates full matrix elements. discuss corresponding general construction criteria show examples gains selected LHC production processes.
We present a study on using Markov Chain Monte Carlo (MCMC) techniques to explore the high-dimensional and multi-modal phase space of scattering events at high-energy particle colliders. To this end, we combine BAT.jl package that provides implementations variety MCMC algorithms with Sherpa event generator framework. discuss technical aspects implementation resulting algorithm first results for process $Z+3$ jets production LHC.
In this talk we report on a novel approach for the integration of scattering cross sections and generation partonic event samples in high-energy physics. It is based an importance sampling algorithm which includes use neural networks order to overcome typical shortcomings conventional approaches. At same time, potential pitfall context phase-space sampling, namely mappings that are non-bijective after trainings with finite data sets, avoided by employing technique Neural Importance Sampling....