# The Tracking Machine Learning challenge

- Institution: CERN
- Department: LPCC Workshops
- Speaker: David Rousseau
- Year: 2019
- Language: English
- Subjects: Computer science
- Source: [CERN](https://videos.cern.ch/record/3008990)
- 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)

The HL-LHC will see ATLAS and CMS see proton bunch collisions reaching track multiplicity up to 10.000 charged tracks per event. Algorithms need to be developed to harness the increased combinatorial complexity. To engage the Computer Science community to contribute new ideas, we have organized a Tracking Machine Learning challenge (TrackML). Participants were provided events with 100k 3D points, and are asked to group the points into tracks; they are also given a 100GB training dataset including the ground truth. The challenge is run in two phases. The first "Accuracy" phase has run on Kaggle platform from May to August 2018; algorithms were judged judged only on a score related to the fraction of correctly assigned hits. The second "Throughput" phase ran Sep 2018 to March 2019 on Codalab, required code submission; algorithms were then ranked by combining accuracy and speed. The first phase has seen 653 participants, with top performers with innovative approaches. The second phase has just finished and featured some astonishingly fast solution. The talk will report on the first lessons from the challenge..

0:00:00 Slide 1

0:00:17 Slide 2

0:00:36 Slide 3

0:00:55 Slide 4

0:01:05 Slide 5

0:01:30 Slide 6

0:02:26 Slide 7

0:03:23 Slide 8

0:04:54 Slide 9

0:05:19 Slide 10

0:05:43 Slide 11

0:06:43 Slide 12

0:07:57 Slide 13

0:08:01 Slide 14

0:08:16 Slide 15

0:08:52 Slide 16

0:09:01 Slide 17

0:09:49 Slide 18

0:10:44 Slide 19

0:10:48 Slide 20

0:11:38 Slide 21

0:12:31 Slide 22

0:12:42 Slide 23

0:13:44 Slide 24

0:13:56 Slide 25

0:14:25 Slide 26

0:14:31 Slide 27

0:14:35 Slide 28

0:14:53 Slide 29

0:15:33 Slide 30

0:15:57 Slide 31

0:17:30 Slide 32

0:18:29 Slide 33

0:18:33 Slide 34

0:19:28 Slide 35

0:20:20 Slide 36

0:20:24 Slide 37

0:20:54 Slide 38

0:21:41 Slide 39

0:22:25 Slide 40

0:22:51 Slide 41

0:24:01 Slide 42

0:25:10 Slide 43

0:25:41 Slide 44

0:25:45 Slide 45

0:26:36 Slide 46

## Lectures

1. [The Tracking Machine Learning challenge](https://invisiblenetwork.ai/learn/cern/3008990/50e2d156d3ca), 33 min (CERN's player)