CERN Accelerating science

ATLAS Note
Report number	ATL-PHYS-PUB-2020-014
Title	Deep Sets based Neural Networks for Impact Parameter Flavour Tagging in ATLAS
Corporate Author(s)	The ATLAS collaboration
Collaboration	ATLAS Collaboration
Publication	2020
Imprint	25 May 2020
Number of pages	20
Note	All figures including auxiliary figures are available at https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/PUBNOTES/ATL-PHYS-PUB-2020-014
Subject category	Particle Physics - Experiment
Accelerator/Facility, Experiment	CERN LHC ; ATLAS
Free keywords	Flavour tagging ; b-tagging ; neural networks ; deep sets ; impact parameter ; dips ; BTAGGING
Abstract	This work introduces a new architecture for Flavour Tagging based on Deep Sets, which models the jet as a set of tracks, in order to identify the experimental signatures of jets containing heavy flavour hadrons using the impact parameters and kinematics of the tracks. This approach is an evolution with respect to the Recurrent Neural Network (RNN) currently adopted in the ATLAS experiment, which treats track collections as a sequence. The Deep Sets model comprises a permutation-invariant and highly parallelisable architecture, leading to a significant decrease in training and evaluation time, and thus allowing for much faster turn-around times for optimisation. Additionally, this permutation invariance encoded in the model is more physically motivated than the sequence-based RNN. We compare the Deep Sets algorithm with the RNN benchmark, probe the model to interpret the information learned, and provide studies optimising the Deep Sets algorithm by loosening the track selection and including additional inputs.
Scientific contact person	Willocq, S, (Stephane.Willocq@cern.ch)
Copyright/License	© 2020-2026 CERN (License: CC-BY-4.0)

Corresponding record in: Inspire