# Identification of Jets Containing b-Hadrons with Recurrent Neural Networks at the ATLAS Experiment

- Institution: CERN
- Department: Machine Learning
- Speaker: Daniel Hay Guest
- Year: 2017
- Language: English
- Subjects: Computer science
- Source: [CERN](https://videos.cern.ch/record/3017179)
- Licence: [CERN](http://copyright.web.cern.ch)
- Who can watch: Tweens profiles and older (sources give no ratings; placed by Invisible Network from the source's own words)

A novel b-jet identification algorithm is constructed with a Recurrent Neural Network (RNN) at the ATLAS Experiment. This talk presents the expected performance of the RNN based b-tagging in simulated $t \bar t$ events. The RNN based b-tagging processes properties of tracks associated to jets which are represented in sequences. In contrast to traditional impact-parameter-based b-tagging algorithms which assume the tracks of jets are independent from each other, RNN based b-tagging can exploit the spatial and kinematic correlations of tracks which are initiated from the same b-hadrons. The neural network nature of the tagging algorithm also allows the flexibility of extending input features to include more track properties than can be effectively used in traditional algorithms..

0:00:00 Slide 1

0:00:11 Slide 2

0:00:15 Slide 3

0:00:35 Slide 4

0:01:32 Slide 5

0:02:19 Slide 6

0:02:58 Slide 7

0:04:04 Slide 8

0:04:42 Slide 9

0:05:40 Slide 10

0:06:55 Slide 11

0:07:25 Slide 12

0:08:19 Slide 13

0:09:11 Slide 14

0:10:00 Slide 15

0:10:47 Slide 16

0:11:33 Slide 17

0:12:40 Slide 18

0:13:22 Slide 19

0:14:14 Slide 20

0:15:48 Slide 21

0:17:46 Slide 22

0:18:56 Slide 23

0:19:01 Slide 24

0:22:52 Slide 25

## Lectures

1. [Identification of Jets Containing b-Hadrons with Recurrent Neural Networks at the ATLAS Experiment](https://invisiblenetwork.ai/learn/cern/3017179/6207c92da0dc), 24 min (CERN's player)