Speaker
Francesco Casini
(Istituto Nazionale di Fisica Nucleare)
Description
In 14 years of operation the Fermi-LAT detected more than 7000 γ-ray sources, of which one third are still not associated with counterparts in other wavelengths, and approximately one fifth are associated with blazar of unknown type. We developed a machine learning method based on an artificial neural network trained with multi-wavelength data which we used to classify blazars of unknown type as either BL Lacs or Flat Spectrum Radio Quasars. Then we considered all the possible multi-wavelength counterparts of the unidentified γ-ray sources, and we implemented another neural network to identify which counterpart was the best candidate and to classify the unidentified sources accordingly.
Primary authors
Francesco Casini
(Istituto Nazionale di Fisica Nucleare)
Sara Cutini
(Istituto Nazionale di Fisica Nucleare)
Stefano Germani
(Istituto Nazionale di Fisica Nucleare)
