Neural Networks and Photometric Redshifts
Roberto Tagliaferri, Giuseppe Longo, Stefano Andreon, Salvatore Capozziello, Ciro Donalek, Gerardo Giordano
Abstract
We present a neural network based approach to the determination of photometric redshift. The method was tested on the Sloan Digital Sky Survey Early Data Release (SDSS-EDR) reaching an accuracy comparable and, in some cases, better than SED template fitting techniques. Different neural networks architecture have been tested and the combination of a Multi Layer Perceptron with 1 hidden layer (22 neurons) operated in a Bayesian framework, with a Self Organizing Map used to estimate the accuracy of the results, turned out to be the most effective. In the best experiment, the implemented network reached an accuracy of 0.020 (interquartile error) in the range 0<zphot<0.3, and of 0.022 in the range 0<zphot<0.5.
Create a lesson
Related papers
On binary pulsars and the force of gravity
Davor Palle
Tidal torques. A critical review of some techniques
Michael Efroimsky, James G. Williams
Dynamics of a Spherical Accretion Shock with Neutrino Heating and Alpha-Particle Recombination
Rodrigo Fernández, Christopher Thompson
Asymptotically FRW black holes
J. T. Firouzjaee, Reza Mansouri
Reaction of Accretion Disks to Abrupt Mass Loss During Binary Black Hole Merger
Sean M. O'Neill, M. Coleman Miller, Tamara Bogdanovic et al.
A Gamma-Ray Burst/Pulsar for Cosmic-Ray Positrons with a Dark Matter-like Spectrum
Kunihito Ioka