72 new bursts, on top of 21 already identified, in five hours of data. Total of 93 bursts in five hours, for an average of one every 3 minutes 14 seconds. That's... rather surprising.
>Initial “standard” algorithms identified 21 FRBs, all happening in one hour’s worth of the observations...Sure enough, the machine learning model picked out 72 more FRBs in the same period.