Speaker
Description
Future gravitational-wave (GW) observatories such as LISA and third-generation ground-based detectors will produce long, high-sensitivity data streams with increased source overlap and weak-signal regimes. We investigate quantum machine-learning approaches for semi-supervised GW anomaly detection, focusing on Quantum Variational Rewinding (QVR) and Quantum Reservoir Computing (QRC). QVR is evaluated on real LIGO data, LISA Data Challenge 2a Sangria data, and synthetic LISA streams with massive-black-hole-binary injections, while QRC is explored as a reservoir-based extension on LISA open data. The results show that quantum anomaly-detection methods can support event flagging, weak-signal screening, anomaly-based ranking, and follow-up prioritization, providing a basis for scaling toward realistic LISA and Einstein Telescope scenarios.
| Topic | Quantum variational circuits and their applications |
|---|