# Applying Generative Models to Scientific Research

- Institution: CERN
- Department: LPCC Workshops
- Speaker: Fedor Ratnikov
- Year: 2019
- Language: English
- Subjects: Computer science
- Source: [CERN](https://videos.cern.ch/record/3016898)
- 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)

Surrogate generative models demonstrate extraordinary progress in current years. Although most applications are dedicated to image generation and similar commercial goals, this approach is also very promising for natural sciences, especially for tasks like fast event simulation in HEP experiments. However, application of such generative models to scientific research implies specific requirements and expectations from these models. In the presentation, I'll discuss specific points which need attention when using generative models for scientific research. This includes ensuring that models satisfy different boundary conditions and match scientifically important but marginal statistics. We also need to establish procedures to evaluate the quality of the particular model, propagate model imperfection into systematic uncertainties of the final scientific result, and so on..

0:00:00 Slide 1

0:00:51 Slide 2

0:01:34 Slide 3

0:04:23 Slide 4

0:06:25 Slide 5

0:07:17 Slide 6

0:07:22 Slide 7

0:07:26 Slide 8

0:07:30 Slide 9

0:07:39 Slide 10

0:07:43 Slide 11

0:07:47 Slide 12

0:07:51 Slide 13

0:08:37 Slide 14

0:09:01 Slide 15

0:10:18 Slide 16

0:10:26 Slide 17

0:10:44 Slide 18

0:11:44 Slide 19

0:12:54 Slide 20

0:13:35 Slide 21

0:16:04 Slide 22

0:18:21 Slide 23

0:20:21 Slide 24

0:20:58 Slide 25

0:22:10 Slide 26

0:23:13 Slide 27

0:24:11 Slide 28

0:25:10 Slide 29

0:25:14 Slide 30

0:26:44 Slide 31

0:27:12 Slide 32

0:27:57 Slide 33

0:28:46 Slide 34

0:28:51 Slide 35

## Lectures

1. [Applying Generative Models to Scientific Research](https://invisiblenetwork.ai/learn/cern/3016898/8c2dedc1e270), 31 min (CERN's player)