# Uncertain Networks

- Institution: CERN
- Department: LPCC Workshops
- Speaker: Jennifer Thompson
- Year: 2019
- Language: English
- Subjects: Computer science
- Source: [CERN](https://videos.cern.ch/record/3008988)
- 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)

Machine learning methods are being increasingly and successfully applied to many different physics problems. However, currently uncertainties in machine learning methods are not modelled well, if at all. In this talk I will discuss how using Bayesian neural networks can give us a handle on uncertainties in machine learning. I will use tagging tops vs. QCD as an example of how these networks are competitive with other neural network taggers with the advantage of providing an event-by-event uncertainty on the classification. I will then further discuss how this uncertainty changes with experimental systematic effects, using pile-up and jet energy scale as examples..

0:00:00 Slide 1

0:00:51 Slide 2

0:01:41 Slide 3

0:02:25 Slide 4

0:02:58 Slide 5

0:03:26 Slide 6

0:03:34 Slide 7

0:04:17 Slide 8

0:05:36 Slide 9

0:08:31 Slide 10

0:09:50 Slide 11

0:10:42 Slide 12

0:12:30 Slide 13

0:14:19 Slide 14

0:15:53 Slide 15

0:16:31 Slide 16

0:17:17 Slide 17

0:18:18 Slide 18

0:18:57 Slide 19

0:20:02 Slide 20

0:21:57 Slide 21

0:22:54 Slide 22

0:24:03 Slide 23

0:24:35 Slide 24

0:24:40 Slide 25

0:25:09 Slide 26

0:27:31 Slide 27

0:28:47 Slide 28

0:29:03 Slide 29

0:30:09 Slide 30

0:31:46 Slide 31

0:31:50 Slide 32

0:31:55 Slide 33

## Lectures

1. [Uncertain Networks](https://invisiblenetwork.ai/learn/cern/3008988/ef26a6d6a3c3), 34 min (CERN's player)