Gravitational waves (GWs) are unique messengers as they travel through the Universe without alteration except for gravitational lensing. Their long wavelengths make them susceptible to diffraction by cosmic structures, providing an unprecedented opportunity to map dark matter substructures. Identifying lensed events requires the analysis of thousands to millions of simulated events to reach high statistical significances. This is computationally prohibitive with standard GW parameter estimation methods. We exploit dingo-lensing, a deep-learning algorithm that accelerates the inference from CPU days to minutes to thoroughly reanalyze GW231123, the most promising lensing candidate to date. By performing more than 200,000 simulations with three different waveform models, we find that its statistical significance is below $4\sigma$ and the event cannot be claimed as lensed. We observe that 8% of GW231123-like nonlensed simulations favor lensing, which could be explained by the self-similarity of short-duration signals. Still, 58% of GW231123-like lensed simulations have larger support for lensing, showing that higher detection statistics are possible. We show that analyzing simulations with different waveform models only lowers the significance, highlighting the relevance of waveform systematics. Although GW231123 exposes the challenges of claiming the first GW lensing detection, our deep-learning methods have demonstrated to be powerful enough to enable the upcoming discovery of lensed GWs.

Close

PrevNext

View figure in article

• Gravitational lenses
• Gravitational waves
• Deep learning