Bright key facts / 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.
- AI’s role
- A reinforcement-learning policy was trained with frequency-domain rewards to suppress control noise while stabilizing the mirror system.
- 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.
- Important limitation
- The projected benefit of hundreds more events per year depends on applying the method across all LIGO mirror-control loops.