Tools for real-time inferences in GW detectors

          @misc{ scivideos_oai:cds.cern.ch:3025629,
            doi = {},
            url = {https://videos.cern.ch/record/3025629},
            author = {},
            keywords = {},
            language = {en},
            title = {Tools for real-time inferences in GW detectors},
            publisher = {},
            year = {2026},
            month = {may},
            note = {oai:cds.cern.ch:3025629 see, \url{https://scivideos.org/cern-cds/3025629}}
          }
          

Abstract

Machine learning will play a key role in the next generation of interferometers data acquisition architectures, particularly through hardware-based solutions. However, methods deployed will have to meet very specific simplicity and robustness requirements. We will present those constraints and the tools we are currently developing to fulfill them at different hardware stages: * **TolmNet**: a C program included in the Virgo online software framework to perform neural network inference in real-time computing environment. * **pyML2FPGA**: a python framework based on an HDL library converting a network into a ressources/latency optimized VDHL code which can be ported to a low-end FPGA. In both cases, we present the results obtained so far using as an input a simple gravitational wave detection network.

00:00:00 Slide 1
00:00:26 Slide 2
00:00:54 Slide 3
00:01:34 Slide 4
00:02:20 Slide 5
00:03:23 Slide 6
00:04:05 Slide 7
00:05:00 Slide 8
00:06:34 Slide 9
00:07:46 Slide 10
00:08:53 Slide 11
00:10:33 Slide 12
00:10:57 Slide 13
00:11:18 Slide 14
00:11:42 Slide 15
00:12:14 Slide 16
00:12:44 Slide 17
00:13:00 Slide 18
00:13:27 Slide 19
00:13:52 Slide 20
00:14:19 Slide 21
00:14:49 Slide 22
00:15:26 Slide 23
00:16:15 Slide 24
00:17:26 Slide 25
00:18:10 Slide 26
00:18:43 Slide 27
00:19:36 Slide 28
00:20:02 Slide 29
00:20:50 Slide 30
00:21:43 Slide 31