Machine learning challenges in theoretical HEP

@article{Carrazza2017MachineLC,
  title={Machine learning challenges in theoretical HEP},
  author={Stefano Carrazza},
  journal={Journal of Physics: Conference Series},
  year={2017},
  volume={1085}
}
  • S. Carrazza
  • Published 29 November 2017
  • Computer Science
  • Journal of Physics: Conference Series
In these proceedings we perform a brief review of machine learning (ML) applications in theoretical High Energy Physics (HEP-TH). We start the discussion by defining and then classifying machine learning tasks in theoretical HEP. We then discuss some of the most popular and recent published approaches with focus on a relevant case study topic: the determination of parton distribution functions (PDFs) and related tools. Finally, we provide an outlook about future applications and developments… 

Machine Learning tools for global PDF fits

The machine learning tools used in the NNPDF family of global QCD analyses are reviewed, including multi-layer feed-forward neural networks for the model-independent parametrisation of parton distributions and fragmentation functions, genetic and covariance matrix adaptation algorithms for training and optimisation, and closure testing for the systematic validation of the fitting methodology.

Quantum Machine Learning in High Energy Physics: the Future Prospects

The present introductory article delivers the future possibilities for the lucid application of quantum machine learning in HEP, rather than focusing on deep mathematical structures of techniques arise in this domain.

Quantum Machine Learning concepts for Physicists

This framework shows the main tools to explore the Standard Model extensions, decay process and the parameter space, with this set of tools, the bounds and exclusion regions are explored.

AI Reloaded: ACAT2017 Conference Summary

  • P. Bhat
  • Physics
    Journal of Physics: Conference Series
  • 2018
The first edition of the Advanced Computing and Analysis Techniques in Physics Research (ACAT) workshop series was held at Fermilab in the year 2000, which was in fact the 7th meeting of its parent

Quantum machine learning and its supremacy in high energy physics

The future prospects of quantum algorithms in high energy physics (HEP) are revealed and particle identification, knowing their properties and characteristics is a challenging problem in experim...

Precision Physics in Extensions of the Standard Model

The Standard Model of particle physics has been remarkably successful at explaining the behaviour of nature at very high energies. It has been thoroughly tested by experiments at the Large Hadron

Invisible Higgs search through vector boson fusion: a deep learning approach

Vector boson fusion proposed initially as an alternative channel for finding heavy Higgs has now established itself as a crucial search scheme to probe different properties of the Higgs boson or for

Components of ML

  • A. Jung
  • Machine Learning: Foundations, Methodologies, and Applications
  • 2022

Machine Learning: Basic Principles

After formalizing the main building blocks of an ML problem, some popular algorithmic design patterns for ML methods are discussed and some main concepts of machine learning are introduced.

References

SHOWING 1-10 OF 44 REFERENCES

Towards the compression of parton densities through machine learning algorithms

The strategy adopted by the PDF4LHC15 recommendation is summarized and a new approach to Monte Carlo PDF compression based on clustering through machine learning algorithms is discussed about.

Modeling NNLO Jet Corrections with Neural Networks

A preliminary strategy for modeling multidimensional distributions through neural networks by considering as input data the two-dimensional next-to-next leading order (NNLO) jet k-factors distribution for the ATLAS 7 TeV 2011 data is presented.

Parton distributions from high-precision collider data

We present a new set of parton distributions, NNPDF3.1, which updates NNPDF3.0, the first global set of PDFs determined using a methodology validated by a closure test. The update is motivated by

PYTHIA 6.2: Physics and manual

The PYTHIA program can be used to generate high-energy-physics `events', i.e. sets of outgoing particles produced in the interactions between two incoming particles. The objective is to provide as

A posteriori inclusion of parton density functions in NLO QCD final-state calculations at hadron colliders: the APPLGRID project

A method to facilitate the consistent inclusion of cross-section measurements based on complex final-states from HERA, TEVATRON and the LHC in proton parton density function (PDF) fits has been

Specialized minimal PDFs for optimized LHC calculations

The method is illustrated by producing SM-PDFs tailored to Higgs, top-quark pair, and electroweak gauge boson physics, and it is determined that, when the PDF4LHC15 combined set is used as the prior, around 11 Hessian eigenvectors are enough to fully describe the corresponding processes.

Efficient Monte Carlo Integration Using Boosted Decision Trees and Generative Deep Neural Networks

New machine learning based algorithms have been developed and tested for Monte Carlo integration based on generative Boosted Decision Trees and Deep Neural Networks. Both of these algorithms exhibit

Parton distributions in the LHC era: MMHT 2014 PDFs

These MMHT2014 PDFs supersede the ‘MSTW2008’ parton sets, but they are obtained within the same basic framework and include a variety of new data sets, from the LHC, updated Tevatron data and the HERA combined H1 and ZEUS data on the total and charm structure functions.

Parton distributions for the LHC run II

A bstractWe present NNPDF3.0, the first set of parton distribution functions (PDFs) determined with a methodology validated by a closure test. NNPDF3.0 uses a global dataset including HERA-II

Minimisation strategies for the determination of parton density functions

We discuss the current minimisation strategies adopted by research projects involving the determination of parton distribution functions (PDFs) and fragmentation functions (FFs) through the training