# Complexity and Random Matrix Approximations

- Institution: CERN
- Department: Conferences & Workshops
- Speaker: James Halverson
- Year: 2019
- Language: English
- Source: [CERN](https://videos.cern.ch/record/3018137)
- Licence: [CERN](http://copyright.web.cern.ch)
- Who can watch: Teens profiles and older (sources give no ratings; placed by Invisible Network from the source's own words)

Efforts to better understand the landscape and swampland can be stifled by computational complexity. I will discuss ways in which complexity could be overcome by learning random matrix approximations to string data, including both opportunities and caveats. As a concrete example, generative adversarial networks will be used to learn random matrix approximations to certain Calabi-Yau data, and physical implications will be discussed..

0:00:00 Slide 1

0:02:39 Slide 2

0:03:40 Slide 3

0:05:27 Slide 4

0:05:56 Slide 5

0:06:19 Slide 6

0:07:41 Slide 7

0:10:40 Slide 8

0:11:18 Slide 9

0:12:36 Slide 10

0:14:14 Slide 11

0:15:59 Slide 12

0:17:26 Slide 13

0:18:18 Slide 14

0:18:31 Slide 15

0:19:03 Slide 16

0:19:53 Slide 17

0:20:33 Slide 18

0:21:05 Slide 19

0:22:12 Slide 20

0:23:54 Slide 21

0:24:16 Slide 22

0:24:45 Slide 23

0:25:40 Slide 24

0:25:48 Slide 25

0:26:08 Slide 26

0:26:18 Slide 27

0:27:19 Slide 28

0:28:11 Slide 29

0:29:00 Slide 30

## Lectures

1. [Complexity and Random Matrix Approximations](https://invisiblenetwork.ai/learn/cern/3018137/d97b07fb3b1f), 30 min (CERN's player)