2021
DOI: 10.1140/epjc/s10052-021-08897-0
|View full text |Cite
|
Sign up to set email alerts
|

Towards a computer vision particle flow

Abstract: In High Energy Physics experiments Particle Flow (PFlow) algorithms are designed to provide an optimal reconstruction of the nature and kinematic properties of the particles produced within the detector acceptance during collisions. At the heart of PFlow algorithms is the ability to distinguish the calorimeter energy deposits of neutral particles from those of charged particles, using the complementary measurements of charged particle tracking devices, to provide a superior measurement of the particle content … Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

0
25
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
9
1

Relationship

1
9

Authors

Journals

citations
Cited by 39 publications
(25 citation statements)
references
References 16 publications
0
25
0
Order By: Relevance
“…Thinking ahead of the possibilities for future algorithmic development, we should consider a global event reconstruction feeding all the particle candidates into a deep neural network that would take care of associating them, without having an intermediate clustering step. An evolution of this approach could result in a different strategy for the optimisation of the detector design as well [19,20].…”
Section: Jets and Resolution Of Hadronic Systemsmentioning
confidence: 99%
“…Thinking ahead of the possibilities for future algorithmic development, we should consider a global event reconstruction feeding all the particle candidates into a deep neural network that would take care of associating them, without having an intermediate clustering step. An evolution of this approach could result in a different strategy for the optimisation of the detector design as well [19,20].…”
Section: Jets and Resolution Of Hadronic Systemsmentioning
confidence: 99%
“…Following this direction may enable further advances, and is expected to lead to new reconstruction paradigms that consider the overall event from the start. Further potential is provided by the increasing power and sophistication of machine learning techniques, which are expected to open new avenues, such as radically different approaches, where the PF candidates are processed directly via ML algorithms to obtain a full event reconstruction [35,36]. Such methods would by-pass the more traditional approaches which rely on the precise reconstruction of the physics objects as an intermediate step.…”
Section: Exploitation For Precision Measurements and Opportunities For Further Developmentmentioning
confidence: 99%
“…With recent developments and particular success of deep learning in high energy physics [15][16][17][18][19], fast simulation models based on a Generative Adversarial Network (GAN) [20] and Variational Auto-Encoders (VAE) [21,22] have emerged (see e.g. [8,[23][24][25][26][27][28][29][30][31][32][33][34][35]).…”
Section: Introductionmentioning
confidence: 99%