ESA researchers use neural network to identify hundreds of rare cosmic anomalies, including several dozen that defy classification
A neural network tool developed by the European Space Agency (ESA) analyzed nearly 100 million cutouts from the Hubble Legacy Archive in just two and a half days, identifying about 1 400 anomalous objects – more than 800 never previously catalogued. Several dozen of these defied classification altogether, showing the potential and the limits of AI in astronomical discovery.










