# Deployable Reinforcement Learning for Interferometric Control

- Institution: CERN
- Speaker: Nikhil Mukund
- Year: 2026
- Language: English
- Source: [CERN](https://videos.cern.ch/record/3025643)
- Licence: [CERN](https://copyright.web.cern)
- Who can watch: Teens profiles and older (sources give no ratings; placed by Invisible Network from the source's own words)

Gravitational-wave detectors present a difficult control problem. They are strongly coupled, nonstationary, and safety-critical, and there is very little room to learn directly on real hardware. In this talk, I will describe a practical approach to reinforcement learning for interferometric sensing and control, drawing on lessons from both kilometer-scale interferometers and tabletop optical testbeds. I will focus on building useful observations from the underlying physics, using scalable simulations and reduced-order models to make training practical, reuse of existing hardware data, and adding supervisory checks for safer online operation. The broader aim is not to replace classical control, but to combine it with search and learning in a way that is more adaptive and more data-efficient for precision optical systems.

00:00:00 Slide 1

00:00:36 Slide 2

00:01:46 Slide 3

00:02:56 Slide 4

00:04:40 Slide 5

00:06:08 Slide 6

00:06:42 Slide 7

00:07:34 Slide 8

00:09:04 Slide 9

00:09:36 Slide 10

00:11:21 Slide 11

00:12:49 Slide 12

00:13:02 Slide 13

00:14:19 Slide 14

00:16:08 Slide 15

00:16:49 Slide 16

00:17:41 Slide 17

00:18:27 Slide 18

00:19:56 Slide 19

00:20:50 Slide 20

00:25:21 Slide 21

00:25:40 Slide 22

00:26:23 Slide 23

00:27:42 Slide 24

00:28:51 Slide 25

00:31:07 Slide 26

00:33:50 Slide 27

## Lectures

1. [Deployable Reinforcement Learning for Interferometric Control](https://invisiblenetwork.ai/learn/cern/3025643/972b70efe6bf), 34 min (CERN's player)