Scalable early warning for binary neutron-star mergers in third-generation gravitational-wave detectors

A deep-learning tour de force led by Martin (former MSc student here, now a PhD student in Belgium) and Rodrigo, on detecting merging binary neutron stars sooner rather than later. Turns out a bit (actually, more than a bit) of convolutional neural networks here and there buys you four orders of magnitude in computational time. For third-generation detectors, that means telling our telescope friends where to look sooner, while the neutron stars are still on their way to merger.

M. Gerini, R. Tenorio, D. Gerosa.
arXiv:2611.10133 [gr-qc].