BRIGHT EVIDENCE PACK / Demonstrated
Quieting the mirrors that listen to space
Researchers developed Deep Loop Shaping, a reinforcement-learning feedback controller tested on the LIGO Livingston Observatory to reduce control noise in a difficult mirror-control loop.
Canonical Bright record · JSON evidence pack · Key-facts embed
Dates and assessment
- Source published
- 2025-09-04
- Bright published
- 2026-09-07
- Substantive update
- None recorded
- Evidence state
- Demonstrated
- Independent verification
- Not established by this source review
- Last source review
- 2026-09-07
The claim in context
The human problem
Tiny vibrations can obscure the faint gravitational-wave signals that observatories are built to measure.
The prior constraint
Traditional feedback control must keep LIGO's mirrors stable without injecting more vibration into the observation band.
AI’s actual role
A reinforcement-learning policy was trained with frequency-domain rewards to suppress control noise while stabilizing the mirror system.
The documented result
The researchers report 30- to 100-fold lower injected control noise than existing controllers in the difficult loop, plus stable repeated operation on the real Livingston hardware.
Why it may matter
Reducing a source of instrument noise may allow scientists to measure fainter or more distant gravitational-wave events.
Limitations
- The projected benefit of hundreds more events per year depends on applying the method across all LIGO mirror-control loops.
- That full-observatory outcome has not been demonstrated in the cited result.
Original evidence
- Using AI to perceive the universe in greater depth · institution
Attribution
Credit Bright AI Future and link the canonical Bright record.
- Link to the canonical Bright record.
- Keep material limitations with the claim they qualify.
- Link to the original evidence when repeating a substantive claim.
- Do not describe a source check or organization-reported result as independent verification.
Linked source material, quotations, trademarks and media remain subject to their owners’ terms. No reuse right is granted for third-party media.