More mysterious bursts of light from the deep space found



[ad_1]

This site may generate affiliate commissions from the links on this page. Terms of use.

There are still many mysteries to unravel in the cosmos, and we could be even closer to solving one of them through machine learning. Researchers from the University of California at Berkeley SETI Research Center have released an AI on Fast Radio Burst (FRB) data from 2017. These momentary flashes can eclipse millions of stars, but we do not do not know what causes them. The new IA-based analysis has detected more than three times more FRB than previous analyzes.

As impressive as this feat is, we still do not know what causes FRB flashes 121102 (or any other FRB source). Some researchers think they have something to do with supernovas or pulsars (rotating neutron stars), but it is also possible that FRBs have a link with extraterrestrial intelligence. That's why SETI is interested in the phenomenon.

The data in question comes from the Green Bank Telescope in West Virginia, which analyzed an object known as FRB 121102 in August 2017. FRB 121102, located in a dwarf galaxy at 3 billion light-years away, it is the only known source of repeat fast radio bursts. All other FRBs detected since their discovery in 2007 have been unique events. A previous analysis of today's data in 2017 showed that 21 FRBs came from FRB 121102, but this may only represent a fraction of what actually exists.

SETI researchers, led by Ph.D. student Gerry Zhang, formed a convolutional neural network to detect fast radio bursts in datasets collected by telescopes. The team calls this the "Breakthrough Listen" project. The technique is not much different at the technical level than using a neural network to identify objects in a photo. After teaching the network what FRBs look like with tagged data, it can then perform accurate assessments with new data.

The only repetitive FRB ever discovered, located in a dwarf galaxy at 3 billion light-years away.

Breakthrough Listen analyzed the 400 terabytes of data from the Green Bank telescope, which contained 21 known FRBs. The AI ​​drew 72 additional flashes, bringing to 93 the number of radio gusts of FRB 121102 in one day.

Having a better FRB activity bookkeeping will help astronomers develop models to explain the signals. By using technologies such as Breakthrough Listen, we can even find that other FRBs are repeated in a way that we have not detected before.

Top Image Credit: NASA / Hubble

Now read: Scientists detect the flash of the fastest and fastest radio. Spaceships with extraterrestrial solar sail can cause fast and mysterious radio explosions. Despite the first reports, SETI does not think

[ad_2]
Source link