Improving cosmological reach of a gravitational wave observatory using Deep Loop Shaping

Improved low-frequency sensitivity of gravitational wave observatories would unlock study of intermediate-mass black hole mergers and binary black hole eccentricity and provide early warnings for multimessenger observations of binary neutron star mergers. Today's mirror stabilization control injects harmful noise, constituting a major obstacle to sensitivity improvements. We eliminated this noise through Deep Loop Shaping, a reinforcement learning method using frequency domain rewards. We proved our methodology on the LIGO Livingston Observatory (LLO). Our controller reduced control noise in the 10- to 30-hertz band by over 30x and up to 100x in subbands, surpassing the design goal motivated by the quantum limit. These results highlight the potential of Deep Loop Shaping to improve current and future gravitational wave observatories and, more broadly, instrumentation and control systems.

Authors

Institutions

Publication Details

Journal
Science
Published
2025-09-04
DOI
https://doi.org/10.1126/science.adw1291
Citations
3
Primary Topic
Pulsars and Gravitational Waves Research
Type
article
Field-Weighted Citation Impact
3.46
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Improving cosmological reach of a gravitational wave observatory using Deep Loop Shaping

Hudson A. Loughlin, B. K. Berger, D. B. Tanner, Markus Wulfmeier et al.
3 citations
Science
Pulsars and Gravitational Waves Research
3.46
article

Improving cosmological reach of a gravitational wave observatory using Deep Loop Shaping

Hudson A. Loughlin, B. K. Berger, D. B. Tanner, Markus Wulfmeier, A. Jennings, M. Landry, T. O’Hanlon, J. W. Richardson, D. Sigg, H. B. Kabagöz, A. S. Ubhi, Abbas Abdolmaleki, Serkan Cabi, A. Ananyeva, D. Griffith, S. Ballmer, K. Kawabe, H. Overmier, Y. Inoue, Michael N. Evans, A. L. James, Tamara Norman, T. Sadecki, Pushmeet Kohli, M. Fyffe, Yuzhu Dong, D. Hoyland, R. Cottingham, G. Vajente, Roland Hafner, Charles Beattie, Katherine M. Ransom, B. Mannix, R. Frey, F. Matichard, T. Guidry, S. Biscans, E. D. Hall, D. Barker, A. Kontos, P. Fulda, George van den Driessche, Weitao Jia, D. E. McClelland, J. Notte, Iain Barr, A. Cumming, E. Schwartz, H. Pham, S. E. Dwyer, Jorge Pullin, Anchal Gupta, E. L. Merilh, H. Y. Huang, R. X. Adhikari, A. Branch, L. Zhang, Hannah Openshaw, Jonas Degrave, J. Glanzer, G. Traylor, O. Patane, M. MacInnis, K. Mason, L. E. Sanchez, N. Aritomi, T. McRae, J. Betzwieser, R. L. Savage, N. A. Holland, E. Capote, S. McCormick, Martin Riedmiller, C. D. Blair, R. K. Crouch, V. Quetschke, N. Kijbunchoo, C. Whittle, M. Fuentes-Garcia, F. Mera, T. Andrić, A. Ejlli, C. N. Makarem, Mitchell B. Robinson, S. K. Apple, Brendan Tracey, R. M. Martin, Hongmin Cao, C. M. Compton, V. Bossilkov, J. A. Giaime, J. S. Kissel, A. Neunzert, Oliver Groth, P. J. Veitch, J. Oberling, L. Sun, C. Cahillane, A. Vibhute, D. D. Brown, B. Weaver, E. Dohmen, R. Gray, J. G. Rollins, B. Willke, T. Shaffer, A. F. Brooks, S. Appert, D. Schaetzl, S. Karki, M. E. Zucker, S. Karat, V. V. Frolov, Craig Donner, D. C. Coyne, E. Bonilla, E. Goetz, J. Gurs, M. G. Schiworski, J. Warner, B. Gateley, J. Harms, S. Gras, Juan Francisco Mandujano Reyes, B. Lantz, Jonas Buchli, Jennifer Bryant, R. Goetz, J. Hanson, K. Komori, J. L. Wright, J. Vanosky, Sandy H. Huang, L. P. Dartez, G. Moreno, P. Thomas, D. J. Ottaway, Leslie Fritz, G. L. Mansell, L. Barsotti, V. A. Xu, M. Ball, M. C. Heintze, B. J. J. Slagmolen, G. McCarrol, W. Frischhertz, M. Lesovsky, I. Abouelfettouh, J. Csizmazia, R. Macas, P. Fritschel, M. Pirello, A. W. Jones, Kazuhiro Yamamoto, R. M. S. Schofield, Andrea Huber, Adrian F. Helmling-Cornell, M. R. Todd, Peter King, D. H. P. Jones, D. Bhattacharjee, F. Llamas Villarreal, D. Davis, J. H. Romie, R. Weiss, Rahul Kumar, M. Laxen, Yu Him Justin Chiu, A. Mullavey, R. Abbott, T. Etzel, R. Schnabel, C. C. Wipf, Christopher Wipf, V. Srivastava, E. J. Sanchez, A. Sanchez, A. Vecchio, J. Hanks, C. Soike, K. D. Giardina, H. Grote, D. Sellers, G. Billingsley, M. Nakano, C. I. Torrie, C. L. Romel, T. J. N. Nelson, W. Parker, N. Bode, A. Pele, Koji Arai, C. Osthelder, R. Mittleman, M. Lormand, E. R. G. von Reis, D. Ganapathy, Jennifer C Driggers, D. Moraru, J. Feicht, R. McCarthy, M. Kasprzack, K. Kuns, R. W. Short, N. Mavalvala, F. Meylahn, S. Soni, M. P. Ross, S. M. Aston, James D. Collins, K. A. Thorne, K. Lee, K. Ryan, F. Clara, N. Maxwell, G. Thornton, M. Thomas, T. Gayer, N. Demos, K. E. Ramire, Matthias Lochbrunner, A. Effer
article en
3 citations

Abstract

Improved low-frequency sensitivity of gravitational wave observatories would unlock study of intermediate-mass black hole mergers and binary black hole eccentricity and provide early warnings for multimessenger observations of binary neutron star mergers. Today's mirror stabilization control injects harmful noise, constituting a major obstacle to sensitivity improvements. We eliminated this noise through Deep Loop Shaping, a reinforcement learning method using frequency domain rewards. We proved our methodology on the LIGO Livingston Observatory (LLO). Our controller reduced control noise in the 10- to 30-hertz band by over 30x and up to 100x in subbands, surpassing the design goal motivated by the quantum limit. These results highlight the potential of Deep Loop Shaping to improve current and future gravitational wave observatories and, more broadly, instrumentation and control systems.

ScienceVol. 389(6764)
University of California, Riverside (US), Leibniz University Hannover (DE), Australian National University (AU), Louisiana State University (US), California Institute of Technology (US), Birmingham City University (GB), University of British Columbia (CA), Universität Hamburg (DE), Montclair State University (US), Kenyon College (US), The University of Western Australia (AU), University of Oregon (US), University of Washington (US), Missouri University of Science and Technology (US), National Central University (TW), LIGO Scientific Collaboration (US), Istituto Nazionale di Fisica Nucleare, Laboratori Nazionali del Gran Sasso (IT), The University of Texas Rio Grande Valley (US), University College Birmingham (GB), University of Florida (US), Google DeepMind (United Kingdom) (GB), Google (United Kingdom) (GB), ARC Centre of Excellence for Gravitational Wave Discovery (AU), Max Planck Institute for Gravitational Physics (DE), Gran Sasso Science Institute (IT), The University of Adelaide (AU), University of Portsmouth (GB), Massachusetts Institute of Technology (US), Syracuse University (US), The University of Tokyo (JP), University of Glasgow (GB), Cardiff University (GB), University of Birmingham (GB), Sungkyunkwan University (KR), Vrije Universiteit Amsterdam (NL), Stanford University (US)
Openalex Percentile: Top 7%
Pulsars and Gravitational Waves Research
3.46
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.