ABCNet: an attention-based method for particle tagging

@article{Mikuni2020ABCNetAA,
  title={ABCNet: an attention-based method for particle tagging},
  author={Vinicius Massami Mikuni and Florencia Canelli},
  journal={European Physical Journal plus},
  year={2020},
  volume={135}
}
  • V. MikuniF. Canelli
  • Published 13 January 2020
  • Physics, Computer Science
  • European Physical Journal plus
In high energy physics, graph-based implementations have the advantage of treating the input data sets in a similar way as they are collected by collider experiments. To expand on this concept, we propose a graph neural network enhanced by attention mechanisms called ABCNet. To exemplify the advantages and flexibility of treating collider data as a point cloud, two physically motivated problems are investigated: quark–gluon discrimination and pileup reduction. The former is an event-by-event… 

Graph neural networks in particle physics

Graph neural networks are trainable functions which operate on graphs—sets of elements and their pairwise relations—and are a central method within the broader field of geometric deep learning.

The LHC Olympics 2020 a community challenge for anomaly detection in high energy physics

An overview of the LHC Olympics 2020 competition, a description of methods deployed in the competition, lessons learned from the experience, and implications for data analyses with future datasets as well as future colliders are reviewed.

Unsupervised clustering for collider physics

We propose a new method for Unsupervised clustering in particle physics named UCluster, where information in the embedding space created by a neural network is used to categorise collision events

UCluster: Unsupervised clustering for collider physics

We propose a new method for Unsupervised clustering in particle physics named UCluster, where information in the embedding space created by a neural network is used to categorise collision events

Semi-supervised graph neural networks for pileup noise removal

  • Tianchun LiShikun Liu Pan Li
  • Computer Science
    The European Physical Journal C
  • 2023
This study implements a semi-supervised graph neural network for particle-level pileup noise removal, by identifying individual particles produced from pileup, and opens up a new direction of fully data-driven machine learning pileup mitigation studies.

Comparing Point Cloud Strategies for Collider Event Classification

This paper compares several event classification architectures based on the point cloud representation of collider events to find the best balance of performance and computational cost for simple pairwise architectures, which are based on learned edge features.

Do graph neural networks learn traditional jet substructure?

The decision-making process for one state-of-the-art graph neural network, ParticleNet, is explored by looking for relevant edge connections identified using the layerwise-relevance propagation technique.

Multimodal learning with graphs

This work introduces a blueprint for multimodal graph learning, uses it to study existing methods and provide guidelines to design new models, and introduces a categorization of graphs as image-intensive, knowledge-grounded and language-intensive models.

Multimodal Representation Learning on Graphs

This work surveys 142 studies in graph AI and puts forward an algorithmic blueprint for multimodal graph AI methods, which is used to study existing methods and standardize the design of future methods for highly complex systems.

References

SHOWING 1-10 OF 32 REFERENCES

Energy flow networks: deep sets for particle jets

Adapting and specializing the “Deep Sets” framework to particle physics, Energy Flow Networks are introduced, which respect infrared and collinear safety by construction and also develop Particle Flow Networks, which allow for general energy dependence and the inclusion of additional particle-level information such as charge and flavor.

and P

    JEDI-net: a jet identification algorithm based on interaction networks

    We investigate the performance of a jet identification algorithm based on interaction networks (JEDI-net) to identify all-hadronic decays of high-momentum heavy particles produced at the LHC and

    Probing stop pair production at the LHC with graph neural networks

    This paper proposes to represent collision events as event graphs and use the message passing neutral network (MPNN) to analyze the events, and finds that the MPNN can efficiently discriminate the signal and back-ground events.

    JEDI-net: a jet identification algorithm based on interaction networks

    We investigate the performance of a jet identification algorithm based on interaction networks (JEDI-net) to identify all-hadronic decays of high-momentum heavy particles produced at the LHC and

    Review of Particle Physics

    The Review summarizes much of particle physics and cosmology. Using data from previous editions, plus 3,062 new measurements from 721 papers, we list, evaluate, and average measured properties of

    The ATLAS Experiment at the CERN Large Hadron Collider

    This paper describes the ATLAS experiment as installed in i ts experimental cavern at point 1 at CERN. It also presents a brief overview of the expec ted performance of the detector.

    LHC Machine

    The Large Hadron Collider (LHC) at CERN near Geneva is the world's newest and most powerful tool for Particle Physics research. It is designed to collide proton beams with a centre-of-mass energy of

    ParticleNet: Jet Tagging via Particle Clouds

    This work proposes ParticleNet, a customized neural network architecture using Dynamic Graph Convolutional Neural Network for jet tagging problems that achieves state-of-the-art performance on two representative jet tagging benchmarks and is improved significantly over existing methods.