The neural network of the detected signals mysterious source
Machine learning is used in many fields, including the SETI project to search for extraterrestrial intelligence. Researchers at the University of California at Berkeley have used it in the framework of the project Breakthrough Listen to study the mysterious signals from a source located at a distance of 3 billion light years from Earth.
signals are radio bursts lasting a few milliseconds. Their source is not exactly known, but it is assumed that they are emitted by distant galaxies. One theory is that outbreaks occur in the explosion of highly magnetized neutron stars located near the supermassive black hole. Also, it is believed that it is the signals of distant civilizations.
Neural Network Breakthrough Listen studied 700 TB of data collected by the radio telescope in Green Bank 26 August 2017. It is revealed from flash 72 galaxy FRB 121102 and former methods showed only 21 signal. The basis of the algorithm Breakthrough Listen became convolution neural network is commonly used for object recognition in images. The project is looking for outbreaks of noise.
The project - just the beginning of the use of neural networks for signal detection. We hope that our success will inspire other researchers to use machine learning astronomy. - announced the project participant Jerry Zhang (Gerry Zhang).
After studying the frequency of outbreaks, scientists have found that signal FRB 121102 no logical order. Even if it does not communicate with other civilizations, further study will help expand the scope of understanding of the universe.
What is the theory of the origin of mysterious signals you practice? Their opinions can be expressed in our Telegram chat.