# Learning New Physics from a machine

- Institution: CERN
- Department: Machine Learning
- Speaker: Andrea Wulzer
- Year: 2018
- Language: English
- Subjects: Computer science
- Source: [CERN](https://videos.cern.ch/record/3006121)
- 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)

We propose using neural networks to detect data departures from a given reference model, with no prior bias on the nature of the new physics responsible for the discrepancy. The model-independent nature of our approach, and its ability to deal with rare signals such as those expected at the LHC, is quantitatively assessed in toy examples..

0:00:00 Slide 1

0:01:04 Slide 2

0:01:30 Slide 3

0:02:07 Slide 4

0:02:18 Slide 5

0:02:35 Slide 6

0:02:47 Slide 7

0:02:57 Slide 8

0:03:44 Slide 9

0:04:20 Slide 10

0:05:11 Slide 11

0:05:45 Slide 12

0:06:11 Slide 13

0:06:34 Slide 14

0:06:50 Slide 15

0:07:19 Slide 16

0:08:29 Slide 17

0:09:27 Slide 18

0:10:15 Slide 19

0:10:36 Slide 20

0:11:54 Slide 21

0:12:01 Slide 22

0:12:15 Slide 23

0:13:03 Slide 24

0:13:24 Slide 25

0:14:14 Slide 26

0:14:52 Slide 27

0:14:56 Slide 28

0:15:09 Slide 29

0:16:08 Slide 30

0:17:14 Slide 31

0:17:58 Slide 32

0:19:50 Slide 33

0:20:41 Slide 34

0:21:20 Slide 35

0:21:41 Slide 36

0:22:06 Slide 37

0:22:46 Slide 38

0:22:51 Slide 39

0:23:26 Slide 40

0:23:34 Slide 41

0:23:50 Slide 42

0:23:59 Slide 43

0:25:10 Slide 44

0:25:36 Slide 45

0:25:43 Slide 46

0:25:52 Slide 47

0:26:30 Slide 48

0:26:44 Slide 49

0:28:06 Slide 50

0:28:39 Slide 51

0:29:08 Slide 52

0:30:06 Slide 53

0:30:21 Slide 54

0:30:26 Slide 55

0:33:03 Slide 56

0:33:12 Slide 57

0:33:16 Slide 58

0:33:37 Slide 59

0:35:42 Slide 60

0:36:43 Slide 61

0:36:47 Slide 62

0:36:53 Slide 63

0:37:26 Slide 64

0:39:55 Slide 65

0:39:59 Slide 66

0:40:03 Slide 67

## Lectures

1. [Learning New Physics from a machine](https://invisiblenetwork.ai/learn/cern/3006121/c15a159dbeca), 42 min (CERN's player)