Enabling Multi Messenger Astronomy with a low-latency LISA data pipeline

In recent years, following the first detection of Gravitational Waves (GWs), we have witnessed the birth of GW Astronomy. So far, there have been more than 50 events recorded, providing us with invaluable information about the nature of the merging binaries. An exceptional case is the event GW170817, a Neutron Star merger, which was observed with both gravitational and electromagnetic (EM) waves. From a single event alone, by combining both ways of observation, we were able to vastly improve our understanding of such cataclysmic events. In the near future, in particular, in the early 2030s, the ESA Laser Interferometer Space Antenna (LISA) is going to be launched. LISA is a space-borne Gravitational-Wave observatory that, in contrast to the present ground-based detectors, is going to be signal-dominated. The LISA data will give us the unique opportunity to observe the merger of supermassive black hole binary systems, which in combination with the EM observations will enable us to push our knowledge boundaries in astronomy, astrophysics, and cosmology. With EMILIA, we aspire to enable multi-messenger astronomy with LISA, by developing a low-latency data analysis pipeline based on Machine Learning techniques. Our proposed methodology will take into account the source confusion problem of LISA, where monochromatic signals and noise artefacts are going to be classified as such and subtracted from the data. We will then apply a fast semi-analytical algorithm on the residual data, in order to swiftly estimate the sky position and time of coalescence of chirping signals. Such a scheme will enable the synergy of LISA and optical observatories on Earth and in space. A prime example is that of the LISA-Athena missions synergy, which would probe the existence of electromagnetic counterpart of massive black hole mergers and extreme mass ratio inspirals, or phenomena like X-ray flares, disk re-brightening, and relativistic jet formations.