Image Processing, Computer Vision, and Deep Learning: new approaches to the analysis and physics interpretation of LHC events

17th International Workshop on Advanced Computing and Analysis Techniques in Physics Research

物理学;计算机科学

Schwartzman, A.^1 ; Kagan, M.^1 ; Mackey, L.^2 ; Nachman, B.^1 ; De Oliveira, L.^3

SLAC National Accelerator Laboratory, Stanford University, 2575 Sand Hill Road, Menlo Park

CA

94025, United States^1

Department of Statistics, Stanford University, Stanford

CA

94305, United States^2

Institute for Computational and Mathematical Engineering, Stanford University, Stanford

CA

94305, United States^3

关键词: Classification methods;    Classification technique;    Computer vision techniques;    High-level features;    Large Hadron collider LHC;    New approaches;    Particle collision;    State of the art;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/762/1/012035/pdf
DOI  :  10.1088/1742-6596/762/1/012035

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

This review introduces recent developments in the application of image processing, computer vision, and deep neural networks to the analysis and interpretation of particle collision events at the Large Hadron Collider (LHC). The link between LHC data analysis and computer vision techniques relies on the concept of jet-images, building on the notion of a particle physics detector as a digital camera and the particles it measures as images. We show that state-of-the-art image classification techniques based on deep neural network architectures significantly improve the identification of highly boosted electroweak particles with respect to existing methods. Furthermore, we introduce new methods to visualize and interpret the high level features learned by deep neural networks that provide discrimination beyond physics-derived variables, adding a new capability to understand physics and to design more powerful classification methods at the LHC.