Artificial neural networks for cosmic gamma-ray propagation in the universe

We explore the potential of an artificial neural network (ANN) based method intelligence to probe propagation cosmic $\gamma$-ray photons in extragalactic Universe. The journey $\gamma$-rays emitted from a distant source like blazar observer at Earth is impeded by absorption through interaction with background light (EBL), leading electron-positron pair production. This process dominates for gamma ray energy above 10 GeV propagating over cosmological distances. effect attenuation characterized physical quantity called \emph{optical depth}, which strongly depends on photon energy, redshift source, and density EBL photons. estimate optical depth values energies sources redshifts range 0.01 1 using three different most promising models. These estimates are randomly divided into two data sets training testing ANN as inputs output. optimization ANN-performance each model employs standard back-propagation (BP) radial-basis function (RBF) algorithms. performance RBF found be superior BP method. In particular, RBF-ANN 40 neurons hidden layer corresponding proposed Finke et al. (2010) shows best